JP2012529347A - System and method for quantitative evaluation of cardiac electrical events - Google Patents

Abstract

A system and method for characterizing aspects of an electrocardiogram signal is shown and uses the schema of primary and secondary analysis to accurately determine the end timing of a signal wave, such as a falling T wave. In one embodiment, the primary analysis schema relates to comparing the voltage amplitudes of a given sampling window, and the secondary analysis schema relates to comparing the results of the primary analysis of a series of sampling windows. The system can include a processor or microcontroller embedded in a system such as an electrocardiogram hardware system, personal computer, electrophysiological system, and the like.
[Selection] Figure 1

Description

  The present invention relates to the field of medical electronic devices. In particular, the present invention relates to electronic systems, devices and methods for acquiring, processing and displaying diagnostic data for humans and animals such as electrocardiogram data.

  ECG (often referred to as “ECG” or “EKG”) is a globally recognized diagnostic method in cardiology, but the most common mistake in interpreting ECG is the most common interpretation of ECG This approach is based on human memory of waveforms rather than using the basic concept of vector concepts and electrocardiography (Hurst, JW, Clin. Cardiol., 2000 Jan; 23 (1): 4- 13). Another problem with conventional ECG recording is that the ECG does not provide an adequate indication of electrical activity in certain areas of the heart, particularly the posterior areas. The timing of cardiac electrical events and the time interval between two or more such events have diagnostic and clinical significance. However, medical diagnosis and drug development are severely limited by the lack of appropriate ECG measurement tools. Furthermore, analysis of conventional ECG recordings required a significant amount of training and familiarity with reading the recorded waveforms. There have been many attempts to extract additional information from standard 12-lead ECG measurements when measuring the potential distribution on the patient's body surface for diagnostic purposes. These attempts include a new approach to interpretation of measurement signals, with or without the introduction of new measurement points, in addition to the standard 12-lead ECG points.

  One of the oldest approaches, Vector ECG (or “VCG”), includes improvements in spatial aspects to ECG (Frank, E., An Accurate, Linearly Practical System for Spatial Vector Cardio, 13: Circulation 37: , May 1956). Similar to conventional ECG interpretation, VCG uses a dipole approximation of electrical heart activity. The size and direction of the dipole is indicated by a vector that continuously changes during the heartbeat cycle. Instead of displaying the signal waveform from the measurement point (waveform) as in the standard 12-lead ECG, in VCG three derived signals correspond to three orthogonal axes (X, Y, Z). Measurement points are arranged so that these signals are displayed as projections of a vector hodograph on three planes (frontal plane, sagittal plane and horizontal plane). Thus, although VCG represents a step towards the spatial representation of the signal, the cardiologist's spatial imagination was still necessary to interpret the ECG signal, particularly in relation to cardiac anatomy. Furthermore, the time-dependent aspect (ie, signal waveform) is lost with this procedure, and this aspect is very important for ECG interpretation. Although VCG introduces useful elements that are not found within the standard 12-lead ECG, despite VCG's very good and accurate diagnosis of cardiac defects such as myocardial infarction (compared to ECG) Unlike ECG, the main reasons why VCG has not been widely adopted are imperfect spatial representation and time-dependent loss.

  Many attempts have been made to overcome the disadvantages of the VCG method described above. These methods use the same signals as VCG (X, Y, Z), but their signal representation is different from vector hodgraph VCG projections on three planes. “Polar cardiogram” uses Aitoff map projection for the display of three-dimensional vector hodographs (see Sada, T., et al., J Electrocardiol. 1982; 15 (3): 259-64). . “Spherocardiogram” adds vector amplitude information to an Aitoff projection by drawing a circle of variable radius (Niederberger, M., et al., J Electrocardiol. 1977; 10 (4): 341). See -6). “3D VCG” projects a hodograph on one plane (see Morikawa, J., et al., Angiology, 1987; 38 (6): 449-56). “4D ECG” is similar to “3D VCG”, except that all heartbeat cycles are displayed as separate loops and the time variable is superimposed on one of the spatial variables (Morikawa, J., et al., Angiology, 1996; 47: 1101-6). The “Chronotopocardiogram” displays a time map of a series of heartbeat activity projected onto a sphere (Titomir, LI, et al., Int J Biomed Comput 1987; 20 (4): 275-82). None of these VCG modifications are widely accepted in diagnostics, but they have several improvements over VCG.

  ECG mapping is based on measuring signals from many measurement points on the patient's body. The signal is displayed as a map of equipotential lines on the patient's torso (see McMechan, SR, et al., J Electrocardiol. 1995; 28 Suppl: 184-90). This method provides important information on the spatial dependence of the ECG signal. However, the disadvantage of this method is the measurement procedure that is prolonged compared to ECG and the weak link between human body potential map and cardiac anatomy.

  Inverse epicardial mapping involves different methods, all of which use the same signal for input data as the signal used in ECG mapping, all of which solve numerically the so-called ECG inverse problem. (See A. van Osterom, Biomedizinisch Technik., Vol. 42-El, pp. 33-36, 1997). As a result, a potential distribution on the heart is obtained. These methods have not resulted in useful clinical devices.

  Cardiac electrical activity can be detected on the body surface using an electrocardiograph, the most common symptom being the standard 12-lead ECG. A typical ECG signal is shown in FIG. P-wave (2) indicates atrial depolarization and characterizes the onset of what is termed the “PR interval”. QRS complex (4) shows ventricular depolarization, starting at the beginning of QRS after PR portion (5) and ending at a point known as “point J” (6). Ventricular repolarization begins during QRS and extends to the end of the T wave (14), which is referred to as the "T wave end (Tend)" (8). The ST portion (10) extends from point J (6) to the beginning or start of the T wave (12). The T wave (14) extends from the beginning of the T wave (12) to the T wave end point (8). U waves (not shown) are present in some ECGs. If present, it merges with the end of the T wave or immediately follows the T wave.

  Physiologically, the T wave is an ECG sign of a repolarization gradient, i.e., the difference in the degree of repolarization at various points in time between various regions of the heart. T-waves are mainly derived from transmural repolarization gradients (see Yan and Antzelevitch; Circulation 1998; 98: 1928-1936; Antzelevich, J. Cardiovasc Electrophysiol 2003; 14: 1259-1272). The apex and front and back repolarization gradients may also contribute (see Cohen IS, Giles WR, and Noble D; Nature. 1976; 262: 657-661).

  The outer layer of the heart (epicardium) is quickly repolarized, the central myocardium is slowly repolarized, and the inner layer (endocardium) is moderately repolarized, creating a transmural repolarization gradient. Referring again to FIG. 1, during the ST portion (10), all layers are partially repolarized to approximately equal extent, and the ST portion (10) is approximately equipotential. When the epicardial layer goes to resting potential prior to the other two layers, a T wave (14) begins at a location referred to as the “T wave start point (Ton)” (12). At the T wave peak (T peak) (16), epicardial repolarization is complete and the transmural repolarization gradient is at its maximum. Subsequently, endocardial cells begin to go to resting potential, thereby reducing the transmural gradient and initiating the down slope of the T wave.

  Eventually, the M cells repolarize and occupy the second half of the downward slope of the T wave. When all layers are at static potential and the transmural gradient disappears, the T wave is completed at the T wave end (8).

  The QT interval (9) can be estimated from the ECG by measuring the time from the end of the PR portion (5) to the T-wave endpoint (8). Abnormal QT intervals characterize susceptibility to arrhythmias, which are usually life-threatening. Such abnormalities may be related to genetic abnormalities, various acquired heart abnormalities, electrolyte abnormalities, and certain prescription and non-prescription drugs. Increasing the number of drugs has been shown to increase the QT interval and is associated with the cause of arrhythmias. As a result, drug regulatory agencies are further investigating drug-induced abnormalities in cardiac electrical activity. The accuracy and accuracy of individual measurements is critical for clinical diagnosis of heart disease and drug safety assessment. Pharmaceutical regulatory bodies around the world now need detailed information on drug effects on heart rate intervals measured from ECG data (M. Malik, PACE 2004; 27: 1659-1669; Guidance for Industry: E14 Clinical Evaluation). of QT / QTc Interval Proportionation and Proarrhytic Potential for Non-Antiarrhythmic Drugs, http://www.fda.gov/cder/guidence/6922fnl.

  Improved measurement accuracy and accuracy will reduce the risk of clinical mistakes and the amount of documentation required to meet regulatory standards during drug development. This is particularly true for QT interval measurements. The problem of manual QT interval determination is due in part to the choice of guidance. The measured QT interval may vary significantly depending on the ECG induction selected for measurement. Another common problem is finding the T-wave endpoint. This is usually defined as the point where the measured voltage returns to the equipotential reference line. However, T waves are often of low amplitude, morphologically abnormal, fused with the next U wave, or obscured by noise. The same would apply to the J point, P wave, U wave and other important cardiac events.

  Therefore, there is an urgent need for an accurate and reproducible procedure for heart rate interval measurement. The present invention addresses this challenge with a relatively noise-resistant solution to determine the timing of cardiac electrical events.

  One embodiment of the present invention is directed to a system for determining a transition point of a signal wave that includes one or more signals sampled from electrodes operatively connected to one or more tissue structures. A memory element configured to store data related to the wave, and a processor operatively connected to the memory element and configured to access the data and determining a transition point associated with the one or more signal waves. The processor samples the first plurality of points of the signal wave of the first time window, and the first plurality of points at least includes a first time point and a last time point of the first time window. And sampling a second plurality of points of a signal wave of a second time window that is different in time from the first time window, wherein the second plurality of points are points of the first time of the second time window And a first plurality of points including at least a last time point The point values are compared with each other to determine whether or not the in-window pattern rule is violated, and after determining whether or not the in-window pattern rule is violated, the secondary analysis is performed, and the secondary analysis is performed in the second plurality of cases. The values of the points are compared with each other to determine whether the in-window pattern rule is violated.

  The system further comprises a display operatively connected to the processor, the processor further configured to cause the display to display at least one graphic image of the one or more signal waves, each of the one or more signal waves. Including a graphic representation of the transition points associated with each of the. The system processor and memory element can be operatively connected to one or more printed circuit boards, one of which comprises a card having a common housing and an electronic interface bus, or is configured to interface with a personal computer. Yes. The processor and memory element comprise an application specific integrated circuit configured to be incorporated into a parent electronic device. In one embodiment, one of the processor and memory element may comprise an application specific integrated circuit configured to be incorporated into a parent electronic device. The processor and memory element may be operatively connected to operating room electronic devices such as electrophysiological mapping systems, echocardiography systems, and fluoroscopic imaging systems. Such an operative connection is via an Ethernet connection, and the various devices can be configured to communicate with each other via a protocol selected from the group consisting of TCPIP, FTP and HTTP. The processor and memory element comprise a personal computer and / or can be operatively connected to an analog signal acquisition system, which can be operatively connected to one or more electrodes. Such an analog signal acquisition system can be, for example, an electrocardiogram system, an electroencephalogram system, or an electromyogram system. In one embodiment, the processor and memory element are encased in an implantable housing, the memory element can be operatively connected to an external computer system to exchange data with the external computer system either wired or wirelessly. Can be configured. In one embodiment, the analog signal acquisition system comprises an emergency holter monitor.

  The processor can be configured to sample a second plurality of points in a second time window forward at least partially in time from the first time window. The signal wave containing the second plurality of points to be sampled can fall in amplitude versus time, and the transition point of the signal wave can be determined by the processor based on the end of the descent of the falling signal wave. The signal wave containing the second plurality of points to be sampled can rise in amplitude versus time, and the transition point of the signal wave can be determined by the processor based on the end of the rise of the rising signal wave.

  In another embodiment, the processor can be configured to sample a second plurality of points in a second time window that is at least partially in time from the first time window. The signal wave including the second plurality of points to be sampled can fall with amplitude versus retrograde time, and the transition point of the signal wave is determined by the processor based on the end of the fall of the signal wave falling with retrograde time Can do. The signal wave including the second plurality of points to be sampled can rise in amplitude versus retrograde time, and the transition point of the signal wave is determined by the processor based on the end of the rise of the signal wave rising in retrograde time Can do.

  Yet another embodiment of the invention is directed to a method for determining a transition point of a signal wave, the first step of sampling a first plurality of points of a signal wave of a first time window, comprising: A plurality of points including at least a first time point and a last time point in a first time window; and a second time window signal wave having a time different from the first time window. Sampling a plurality of two points, wherein the second plurality of points includes at least a first time point and a last time point of a second time window; and the first plurality of points Comparing the values of each other with each other to determine whether the pattern rule within the window is violated in the first time window, and performing a secondary analysis after determining that the pattern rule within the window is violated A secondary analysis window that compares the second plurality of values to each other And determining whether the pattern rule has been broken in the second time window.

  The secondary analysis further includes determining whether the inter-window pattern rule is violated and associating a signal wave transition point with the position of the signal wave where the inter-window pattern rule is violated. In another embodiment, the method further includes characterizing the noise level of the signal wave. Characterizing the noise level may include fitting a pass curve of data points that include signal waves, such as a polynomial that best fits the data points. Characterizing the noise level further includes determining a change in root mean square of the data points for the curve. In another embodiment, the method may further include selecting an in-window pattern rule based at least in part on the noise level of the signal wave. In another embodiment, the method may further include selecting an inter-window pattern rule based at least in part on the noise level of the signal wave. The in-window pattern rule can be automatically selected based on a computer analysis of the signal wave. Similarly, the window pattern rule can be automatically selected based on computer analysis of the signal wave. The signal wave can include an analog-to-digital converted electrocardiogram signal and is associated with one of a plurality of electrodes operatively connected to the patient.

  In another embodiment, the signal wave can include a vector magnitude T-wave signal representation derived from the electrocardiogram voltage amplitude and is associated with a plurality of electrodes operatively connected to the patient. The signal wave can include a voltage amplitude plotted against time, and a window when the difference between the first time point and the last time point of each of the first plurality of points is greater than a predetermined threshold voltage amplitude difference. It can be considered that the inner pattern rule has been broken. The inter-window pattern rule can be considered broken based at least in part on a pattern that breaks the intra-window pattern rule in the first and second time windows. The secondary analysis can be performed for each of the X projection, the Y projection, and the Z projection including the vector magnitude T wave signal display. The method can further include associating the end of the QT interval with the slowest T wave projection in the time of the X, Y and Z projections of the vector magnitude T wave signal representation. In another embodiment, the method further rotates the X, Y and Z coordinate systems associated with the X, Y and Z projections to time align the X, Y and Z projections of the vector magnitude signal representation in time. Steps may be included.

  In one embodiment, the signal wave can be selected from the group consisting of an electrocardiogram signal, an electroencephalogram signal, and an electromyogram signal. The second time window may be at least partially forward in time from the first time window, and the signal wave is descending in amplitude versus time, and the method further includes at least partially in amplitude of the signal wave. Determining the transition point of the signal wave based on the end of the descent. The second time window may be at least partially forward in time from the first time window and the signal wave is rising in amplitude versus time, and the method further includes at least partially the amplitude of the signal wave. Determining the transition point of the signal wave based on the end of the rise. The transition point of the signal wave is the start of the P wave of the electrocardiogram, the end of the P wave of the electrocardiogram, the end of the PR part of the electrocardiogram, the point Q of the electrocardiogram, the S point of the electrocardiogram, the start of the T wave of the electrocardiogram, the T wave of the electrocardiogram A selection can be made from the group consisting of the end of the wave, the start of the U wave of the electrocardiogram, and the end of the U wave of the electrocardiogram. In another embodiment, the transition point of the signal wave is the start of the ECG P wave, the end of the PR portion of the ECG, the ECG R point, the ECG J point, the start of the ECG T wave, the ECG U wave Can be selected from the group consisting of the start of and the end of the ECG U-wave.

  In one embodiment, the second time window can reverse time at least partially from the first time window, and the signal wave descends in amplitude versus retrograde time, and the method further includes at least partially And a step of determining a transition point of the signal wave based on the end of the fall of the amplitude of the signal wave in the retrograde time. In another embodiment, the second time window can reverse time at least partially from the first time window, and the signal wave rises in amplitude versus retrograde time, and the method further includes at least partly And determining the transition point of the signal wave based on the end of the increase of the amplitude of the signal wave in retrograde time. The transition point of the signal wave is the start of the ECG P wave, the end of the ECG P wave, the ECG Q point, the ECG S point, the ECG J point, the ECG T wave end, the ECG U wave start, And a group consisting of the end of the U wave of the electrocardiogram. In another embodiment, the transition point of the signal wave is the end of the ECG P wave, the end of the PR portion of the ECG, the ECG R point, the ECG J point, the start of the ECG T wave, the ECG T wave. Can be selected from the group consisting of: ending, beginning of ECG U-wave, and ending ECG U-wave.

  In one embodiment, the method further compares each signal wave transition point of the normal population of interest with the determined signal wave transition point and treats the relative position of the determined signal wave transition point. A step of determining whether the application has affected. Such a method may further include changing or stopping the application of the therapy based at least in part on the comparison of the signal wave transition points. This treatment can include, for example, chemotherapy, and the determined signal wave transition point can be the end point of the falling T wave of the ECG signal.

FIG. 1 shows an aspect of a conventional ECG signal. FIG. 2A shows windowed sampling and analysis in a particular aspect of the falling T-wave of the ECG signal. FIG. 2B shows windowed sampling and analysis in a particular aspect of the falling T wave of the ECG signal. FIG. 2C shows windowed sampling and analysis for a particular aspect of the falling T-wave of the ECG signal. FIG. 2D shows windowed sampling and analysis in a particular aspect of the falling T wave of the ECG signal. FIG. 3 shows various aspects of the ECG signal analysis configuration, whereby the QT interval can be measured according to the present invention. FIG. 4 shows various aspects of the ECG signal analysis configuration, whereby the QT interval can be measured according to the present invention. FIG. 5 illustrates various aspects of an ECG signal analysis configuration, whereby the QT interval can be measured in accordance with the present invention. FIG. 6 shows various aspects of the ECG signal analysis configuration, which can be used in analyzing drug effects and managing therapy according to the present invention. FIG. 7 shows various aspects of the ECG signal analysis configuration, whereby the reference position of the signal wave can be compared in the retrograde time analysis according to the present invention. FIG. 8 shows various aspects of a generalized signal wave analysis configuration, whereby signal waves with increasing amplitude and signal waves with decreasing amplitude in the forward and / or backward time according to the present invention. Alternatively, the reference position of the signal wave can be measured for a signal wave whose amplitude is flat and changing. FIG. 9 illustrates an ECG system that can be integrated with embodiments of the present invention. FIG. 10 illustrates an emergency holter monitor system system that can be integrated with embodiments of the present invention. FIG. 11 illustrates an electrophysiological mapping system that can be integrated with embodiments of the present invention. FIG. 12 illustrates an echocardiography system that can be integrated with embodiments of the present invention. FIG. 13A shows a fluoroscopy system that can be integrated with embodiments of the present invention, and FIG. 13B shows a fluoroscopy system that can be integrated with embodiments of the present invention.

  Referring to FIGS. 2A-2D, there is shown an enlarged view of the T wave (14) as shown in FIG. 1, and how T using the data related to the falling T wave signal according to the present invention is shown. It shows whether the wave end point is determined. Referring to FIG. 2A, a portion of the descending T-wave is shown near the voltage amplitude versus time coordinate axis (18). If the illustrated T-wave is actually falling, it is assumed that the point sampled earlier in time has a smaller amplitude, which is the illustrated “sampling window” (20). This applies to the four sampling points (22) shown. In fact, referring to FIG. 2B, the trend continues as the sampling window (20) advances forward in time and captures different points (24), with the amplitude ahead in time being smaller. Referring to FIG. 2C, the sampling window (20) advances time forward again relative to the coordinate axis (18), and the amplitude of the points ahead of this captured points (26) is smaller. Referring now to FIG. 2D, an apparent inflection point of the descending T-wave (14) is reached when the sampling window (20) moves forward again to capture the fourth plurality of points (28). It is apparent from the amplitude of the T wave in the sampling window (20) that at least two of the four sampled sampling points (28) have approximately the same amplitude. In contrast to tangent-based techniques or techniques that rely on finding an intersection with a reference line, in one embodiment of the present invention, the comparison between amplitudes sampled within a given sampling window provides an immediate Is not a good decision, but leads to a second level of analysis. A multi-step approach to find endpoints such as T-wave endpoints using windowed sampling analysis on the descending signal is shown in FIG.

  Referring to FIG. 3, the raw ECG data is sampled at one or more electrodes, such as the conventional 12 sets, converted using a DA converter, and preferably as shown in the signals shown in FIGS. 2A-2D. Yields a set of streams, arrays or points associated with each electrode in units of voltage amplitude versus time (30). Preferably, sampling is performed at a high frequency such as 250-500 Hz. Such high frequency sampling not only provides high-fidelity sampling of voltage amplitude data associated with the electrodes involved, but also provides a high-fidelity representation of noise associated with such signals. As shown in FIG. 3, simultaneous analysis or sequential analysis of noise levels (32) and determination of which ECG signal to use for T-wave end point analysis (34) includes windowed analysis of selected T-waves ( Precedes 36). Noise analysis and signal selection are discussed further below with reference to FIGS. Referring again to FIG. 3, as shown with reference to FIG. 2A, the first plurality of voltage amplitude points can be sampled using the position of the first time window on the selected T-wave. (36). For example, the graphic windows and sampling process shown in FIGS. 2A-2D are useful for illustration, but in one embodiment, such analysis is performed numerically using a computer and raw sample data (30). ). The first plurality of points can be sampled (36), a primary analysis can be performed, and the amplitudes of the points including the sampled points can be compared. In one embodiment, the first time to construct a plurality of points, such as the point labeled “P4” at the plurality of points (22) shown in FIG. 2A (the first time is shown in FIGS. 2A-2D). The amplitude of a point (defined as the rightmost point using the amplitude vs. time axis (18)) is any or all of the adjacent points that make up the points, or more simply, in FIG. The last time constituting a plurality of points, such as the point labeled “P1” at the plurality of points (22) shown (the last time is the amplitude versus time axis (18) shown in FIGS. 2A-2D) Can be compared to the point of the leftmost point). Such a comparison is preferably related to “intra-window” pattern rules (ie, for points in the window of interest), or “threshold”, and whether such rules have been violated. Or a determination as to whether a threshold has been exceeded. For example, in one embodiment, the in-window pattern rule of the primary analysis indicates that the point that is the foremost in time (ie, the rightmost on the time axis (18)) is the farthest in time (ie, the leftmost on the time axis (18)). ) Is defined as having an amplitude smaller than the point of. The amplitude of P4 can simply be subtracted from the amplitude of P1, and a determination is made whether the resulting value is a positive number. In another embodiment, the value obtained to indicate that the rule has been violated or exceeded a threshold need not be negative or zero. For example, in one embodiment, the system is configured to discover that a rule has been breached or exceeded a threshold when P4 is within 10%, 20%, 30%, or some other percentage of the value of P1. be able to.

  In another embodiment, an in-window pattern rule for primary analysis is defined in which the average amplitude of the two points that are the foremost in time at a plurality of points must be less than the average amplitude of the two points that are the foremost in the times of the points To do. The sampling window (20) can be configured to capture a small number of points, such as two points, or more points, such as four points, five points, six points or more.

  In another embodiment, in a polynomial curve that fits through a plurality of points, the slope between the two most forward points at the time of the plurality of points is greater than the slope of the curve between the two most backward points at the time of the plurality of points. The in-window pattern rule of the primary analysis is defined as having a slope that is equal to or less than a specific ratio of positive numbers.

  Referring again to FIG. 3, if the primary analysis rule is violated or exceeds the threshold, the fictitious sampling window (20) is advanced forward in time, as in the difference between FIG. 2B and FIG. As described with respect to FIG. 2B, a second plurality of points is captured (40). Preferably, primary analysis (42) is performed on the second plurality of points using the same in-window pattern analysis as in (38) above. This process of continuing the sampling window and performing the primary analysis can be repeated until the rule is broken or the threshold is exceeded (44), and then a secondary analysis is preferably performed (46). Considering the pattern schema of “inter-window” (ie, one window versus another window), the primary analysis results are compared for adjacent time windows and such rules are violated or thresholded. Decide whether or not. It is preferable that the inter-window rule or threshold value is configured to compare the results of in-window analysis for adjacent windows. For example, in one embodiment where computing resources are readily available, a primary analysis can be performed on the entire descending T-wave data set. The data is then observed from left to right on the time / amplitude axis (18 in FIGS. 2A-2D) and either near the sampling window that caused the failure of all of the sampling windows and the first in-window pattern rule, or directly The failure of the first in-window pattern rule with respect to adjacent points has a secondary analysis between the windows. In one embodiment, more than two directly consecutive intra-window rule failures are associated with an inter-window rule failure. In another embodiment, the failure of more than three primary analysis rules from a directly contiguous group of five time windows is associated with the failure of the inter-window rule. In other words, it is possible to customize the allowable noise range and the failure of the secondary analysis rule, for example, two of the rows, three of the five, five of the seven, and three of the rows. Referring to FIG. 3 again, when the inter-window rule is broken in the secondary analysis, it is possible to determine the QT interval given the T wave end point, that is, the QT start point.

  Referring now to FIG. 4, an embodiment is shown that uses repetition to refine the determination of the T wave endpoint. As shown in FIG. 4, raw ECG data is preferably acquired and stored in memory using a device or system available from GE Medical Systems under the trade name Plucka (RTM) (50). A primary and secondary analysis can then be used to determine a provisional T-wave endpoint based on rule breakdown and data analysis related to applicable intra-window and inter-window rules (52). Given a provisional T-wave end point, further analysis can then be performed and the possibility examined in view of factors such as noise in the raw data (54). For example, in one embodiment, a polynomial curve can be fitted via raw data, and a root mean square (“RMS”) analysis can be performed to quantitatively characterize the error considering the curve. Data when the position of the provisional T-wave end point is too far from what was specified by a simple reference line, tangent, or other analysis performed on the fitted curve, and / or considering RMS analysis In particular, if there is noise, the pattern rules for primary and secondary analysis can be adjusted, and a second provisional T-wave end point is determined (56). For example, if the raw data of a particular T-wave is significantly noisy, a very conservative primary analysis in-window rule may immediately lead to a secondary analysis. In addition, even if appropriately adjusted primary analysis in-window rules are used, or repeated, resulting in secondary analysis at the desired time, the secondary analysis inter-window rules are too conservative or sufficient. If it is not conservative, the determination of the T-wave end point may be too early or too late. Preferably, the computer system is configured to perform such repeated analysis, and such processing is automated by input of a predetermined logic circuit based on experience and experimental data, and pattern rules for primary and / or secondary analysis. It is preferable to provide confirmation (58) in selecting and final T-wave end point determination.

  As described above with reference to FIG. 3, one of the challenges of the raw ECG signal is actually noise, and the potentially noisy data of 12 electrodes by high fidelity and high frequency sampling is of interest. Leaving challenges in determining the patient's exact T-wave endpoint. In one embodiment, noise can be filtered using statistical techniques based on relatively large sample sizes, such as, for example, 24-hour ECG monitoring survey results with emergency hardware such as a Holter monitor. For each electrode, a relatively large sample size signal pattern is available for analysis, e.g. cardiologists interested in drug-related T-wave morphology changes remove statistical outliers from the primary data set Can be recorded for further analysis. In general, however, very large sample sizes cannot be used unless there is a suspicious arrhythmia. In other embodiments, known filtering and smoothing techniques can be used.

  In a further embodiment, a vector magnitude signal is desirable as described above to substantially cancel more noise from the raw ECG guidance data. However, one of the challenges of vector magnitude based analysis is that it relies on accurate data of the zero reference marker at the start of the QT interval. The end of the PR portion (element 5 in FIG. 1) is generally recorded better in the ECG signal than the T-wave end, but if the associated electrode is not fully attached, the signal may be muscle tremor noise or environmental noise. At least partially, making it more difficult to determine the end of the PR portion. In one embodiment based on vector magnitude, the following compromise is made between a fairly noisy analysis based on a relatively large number of electrode signals and the reference point dependence of the vector magnitude signal: their shape is the reference point (The reference point may change such a T wave up and down in amplitude, but does not change their shape), so that each of the X, Y and Z vector components of the T wave is linear And secondary analysis. The result is three relatively accurate measurements of the T wave endpoint for each of the X, Y and Z vector components of the T wave. In one embodiment, the QT interval is determined using the T-wave endpoint farthest from the secondary analysis of the X, Y and Z vector components of the T-wave. In another embodiment, the associated X, Y, Z coordinate system is rotated to generate a T-wave end time simultaneously for each of the T-wave X, Y, and Z vector components (ie, the smallest difference between the three). Rotate the coordinate system to the position where is obtained) and determine one T-wave end point.

  Referring to FIG. 5, another embodiment is illustrated, and in fact, various aspects of analysis can be performed at various times for patient treatment. Referring to FIG. 5, it is preferable to obtain an ECG signal and store it in memory in a pre-operative environment rather than during an operation where all the same changes such as anesthetics and antithrombotic agents are given (62). The primary and secondary analysis (64) to determine the tentative T-wave end point, together with the examination of the tentative T-wave end point taking into account noise and other factors (66), as described above with reference to FIG. Depending on computing resources, patient treatment timing, the need for repetition, or especially the need to infer data from ECG signals in which noise appears, etc. can be done instantaneously or after acquisition. Similarly, repetition of the primary and secondary analysis pattern rules (68) and confirmation of the selected pattern paradigm (70) can be performed after data acquisition or by the remaining patients available for further data acquisition. Such a confirmed paradigm is then saved and utilized in a later scenario of a particular patient, such as a further outpatient visit or surgical procedure (74), to make a convenient and refined determination of the T-wave endpoint. Can be provided (76).

  Referring to FIG. 6, one embodiment is shown where the signal processing paradigm described above with reference to FIGS. 2A-5 can be utilized to assist in the management and explanation of treatment. As shown in FIG. 6, after obtaining preoperative ECG data and storing it in memory (62), the primary and secondary analyzes are performed to end the T wave of a specific patient, that is, the ECG timing such as the T wave end point. A preliminary pre-dose value (ie, pre-dose or pre-treatment for a particular drug) can be determined for the baseline (106). As described above with reference to FIG. 5, provisional values for criteria such as the T-wave end point can be inspected taking into account the noise level of the acquired ECG signal wave (108) and repeated to perform primary and Improve the pattern rules of the secondary analysis, inspect further evolved provisional reference timing values (110), and then confirm the pattern rules selected in the primary and secondary analysis (112) for later comparison Determine the pre-operative reference timing value to be used. After this analysis, in an intraoperative or clinical environment where the patient is operatively connected to the analysis system, further ECG data is acquired and stored in memory (114). The primary and secondary analysis can then be performed on the newly acquired data by applying the established primary and secondary pattern rules (116), and a pre-dose reference timing value such as the timing position of the T wave end point is determined. (118), such a value can be used as a “control” value for later comparison. In another variation, a fixed reference timing value determined using pre-operative data can be used as a “regulation” value for comparison. After applying treatment such as ingestion, injection, or other delivery of one or more chemotherapy or other drugs to the target patient (120), the selected primary and secondary pattern rules for primary and secondary analysis Applied, a reference position during dosing (ie, during treatment or after a specific dose of drug), such as a T-wave end point timing value, can be determined (122). Medication values are those of the population, such as preoperative values, and can be compared to preoperative values of the same patient rather than preoperatively, as well as data values for a population of selected healthy human populations. Can be compared (124) and the resulting comparison information can be used by the physician to change, stop, or otherwise affect the ongoing treatment (126). Thereafter, further ECG data can be acquired to monitor subsequent symptoms and observed after any medication (ie, a predetermined time after the previous drug administration) (128).

  It is important to note that the primary and secondary analysis techniques described in this document are widely applicable. The previously discussed scenario involves a window analysis of forward time (ie, the direction of an event such as an ECG signal occurring in real time), and T associated with the end of a signal wave of decreasing amplitude, such as a T wave. Determine clinically relevant reference timing positions such as wave endpoints. This primary and secondary analysis technique can also be applied to rising and falling signal waves, and flat signal waves, both forward and backward times, for a given signal wave or set of signals. Various reference positions of interest can be located. FIG. 7 shows an embodiment of retrograde in time, or “retrograde time”, and FIG. 8 is a more generalized implementation for applicability over ECG signal waves, ie applicability to any kind of signal wave. The form is shown.

  Referring to FIG. 7, an ECG signal is acquired and converted (30) as described with reference to FIG. 3, and the noise level is analyzed (32) and selected (34). Subsequently, the selected signal wave is sampled in the first window (130), the primary analysis is performed (132), and then the primary analysis of the further windows (134, 136, 138) is continued, Look at the signal wave data and continuously move the sampling window backwards in time along the signal trace (e.g., if plotted with Cartesian amplitude vs. time as in FIGS. 2A-2D, go left) Or go backwards in time and move the sampling window). Once the intra-window pattern rule is breached, a secondary analysis of the retrograde time scenario is performed to determine whether the inter-window pattern rule is breached (140), and then the selected reference timing position is used as the inter-window pattern rule. Can be determined based on the failure (142). In one embodiment, it is desirable to only perform forward time analysis, as described with respect to FIGS. 2A-6. In one embodiment, it is desirable to perform only retrograde time analysis, as described herein with reference to FIG. In another embodiment, it is desirable to perform both forward time and retrograde time analysis. Using these techniques, the timing position of the start of the P wave, the apex of the P wave, the start of the PR portion, the end of the PR portion, the start of the Q point, the R point, the S point, and the ST portion Any ECG reference position can be determined, such as the end of the ST portion, the apex of the T wave, the end of the T wave, the start, apex, and end of any U wave.

  As discussed above, other human electronic signal traces such as electroencephalogram (“EEG”) signals, electromyogram (“EMG”) signals, and other biological and non-biological signal waves, etc. This ingenious primary and secondary analysis can be applied to any signal wave or trace. A generalized embodiment is shown in FIG. Referring to FIG. 8, an analog signal is acquired using one or more electrodes and converted to digital (144), the subsequent noise level is analyzed (146), and the target signal is determined (148). A primary analysis of the first sampling window of the target signal wave can be performed, which is a signal wave that falls in the forward or backward time (150, 152) (the T wave signal wave part that falls when considered in the forward time) The position of the rising signal wave, the flat signal wave, and repeating this analysis, moving the series of sampling windows until the primary pattern window is broken (154, 156, 158), after which A next analysis can be performed (160) and a reference position is determined (162). Optionally, the opportunity to further perform a similar analysis in the opposite time window can be used to compare results in different directions and be used to determine the final reference position.

  In practice, the techniques described with reference to FIGS. 2A-8 are based on customized software, such as semi-customized software based on spreadsheets or software packages available under the trade name LabView (RTM) from National Instruments. One or more computer systems, such as personal computers, that utilize a customized configuration with applications such as, and / or hardware configured to run embedded software. In some embodiments, the related systems are integrated electronically to facilitate real-time or near real-time analysis according to the techniques described above. For example, referring to FIG. 9, in one embodiment, the ECG acquisition system (78) and associated electrodes (80) are preferably integrated into the computer (100) using a wired or wireless connection (84). The computer (100) receives and / or requests data from the ECG system (78) and includes a card (in one embodiment, housed in a card housing and including an electromechanical card interface that includes an integrated circuit and / or memory. Control and / or receive information on the activation of embedded devices (88), such as system-connected buses, application specific integrated circuits ("ASICs"), or field programmable gate arrays ("FPGAs") Each can receive from the computer (100) Preferably, the computer is configured to perform a primary and / or secondary analysis of the raw data received by the ECG system (78) forming the electrode (80) according to a desired command or control sequence, and the computer is at the time of sampling or sampling Connect before. Referring to FIG. 10, an emergency and portable Holter ECG system (88) is similarly connected to an embedded device (82), which is operatively connected from such an electrode set (86) to such a system ( 88) is configured to perform primary and / or secondary analysis based on the raw data received. A bus or connector (90) can be provided for connection to a computer system (not shown).

  Referring to FIGS. 11-13B, in accordance with the present invention, other medical information processing systems generally associated with ECG signal processing are preferably integrated with or incorporated into the primary and secondary processing infrastructure. For example, referring to FIG. 11, an electrophysiological mapping system (92) can also be operatively connected to an embedded device (82), such as that available from Biosense Webster under the trade name CartoXP (RTM). Is configured to perform primary and / or secondary analysis based on raw data received by such a system (92) from an electrode set (not shown) operatively connected to an electrode connection bus panel (94). Is done. The T wave endpoint and other results can be displayed on one or more displays (96). Referring to FIG. 12, operatively connecting an echocardiography system (98) to a computer system (100) and an ECG system (78), such as those available from Siemens Medical Systems, Inc. under the trade name Sequoia (RTM). Can do. An embedded device (82) configured to perform primary and / or secondary analysis based on raw data received from the ECG system (78) includes an ECG system (78), computer system (100) as shown in FIG. Or any one of the echocardiography systems (98). Data relating to primary and secondary analysis is preferably displayed on either an echocardiographic display (96) or a computer system display (97). Similarly, referring to FIGS. 13A and 13B, a relatively simple x-ray fluoroscopy system (102) as shown in FIG. 13A, or a more complex angiography system (104) as shown in FIG. 100), operating on an associated ECG system (78), embedded device, or device configured to perform primary and / or secondary analysis based on raw data received at electrodes operatively connected to other systems Can be connected to each other and / or incorporated therein. Connections of various components of such system configurations, such as processors, memory elements, and operating room electronic devices, are made using Ethernet, wireless technology, and / or communication protocols such as TCPIP, FTP, or HTTP. It can be carried out.

Claims (42)

In a method for determining a transition point of a signal wave,
a. Sampling a first plurality of points of a signal wave of a first time window, wherein the first plurality of points are a first time point and a last time point of the first time window; Including at least a step,
b. Sampling the second plurality of points of the signal wave in a second time window that is different in time from the first time window, wherein the second plurality of points are defined in the second time window; Including at least a first time point and a last time point;
c. Comparing the values of the first plurality of points with each other to determine whether an in-window pattern rule is violated in the first time window;
d. Performing a secondary analysis after determining that the in-window pattern rule has been violated, wherein the secondary analysis compares the second plurality of values with each other and the in-window pattern rule is in the second time period. Determining whether it has been broken by a window.   The method of claim 1, wherein the secondary analysis further comprises determining whether an inter-window pattern rule is violated.   3. The method of claim 2, wherein the inter-window pattern rule is considered to be broken based at least in part on a pattern that breaks an in-window pattern rule in the first and second time windows. Method.   4. The method according to claim 2, further comprising associating a signal wave transition point with the position of the signal wave where the inter-window pattern rule is broken.   The method according to any one of claims 2 to 4, further comprising the step of selecting an inter-window pattern rule based at least in part on the noise level of the signal wave.   5. The method according to claim 2, wherein the window pattern rule is automatically selected based on a computer analysis of the signal wave.   7. The method according to claim 1, further comprising characterizing a noise level of the signal wave by fitting a pass curve of data points including the signal wave. .   8. The method of claim 7, wherein fitting a curve includes fitting a polynomial that best fits the data points, and characterizing a noise level further comprises a root mean square of the data points for the curve. A method comprising determining a change.   9. The method according to any one of claims 1 to 8, further comprising selecting an in-window pattern rule based at least in part on a noise level of the signal wave.   9. The method according to claim 1, wherein the window pattern rule is automatically selected based on a computer analysis of the signal wave.   11. The method according to any one of claims 1 to 10, wherein the signal wave comprises an analog-to-digital converted electrocardiogram signal and relates to one of a plurality of electrodes operatively connected to a patient. Feature method.   11. A method according to any one of the preceding claims, wherein the signal wave comprises a vector magnitude T-wave signal representation derived from an electrocardiogram voltage amplitude and is associated with a plurality of electrodes operatively connected to a patient. The method is characterized in that the secondary analysis is performed for each of an X projection, a Y projection and a Z projection including the vector magnitude T-wave signal representation.   13. The method of claim 12, further comprising associating an end of a QT interval with a T-wave projection that is slowest in the time of X, Y, and Z projections of the vector magnitude T-wave signal representation.   The method of claim 12, further comprising rotating the X, Y and Z coordinate systems associated with the X, Y and Z projections to align the T-wave endings of the X, Y and Z projections of the vector magnitude signal representation in time. A method comprising the steps of:   11. A method according to any one of the preceding claims, wherein the signal wave comprises a voltage amplitude plotted against time, each of the first and last time points of the first plurality of points. And determining that the in-window pattern rule has been violated when the difference in voltage is greater than a voltage amplitude difference of a predetermined threshold.   The method according to any one of claims 1 to 10, wherein the signal wave is selected from the group consisting of an electrocardiogram signal, an electroencephalogram signal and an electromyogram signal.   17. The method of claim 16, wherein the second time window is at least partially forward in time from the first time window, and the signal wave is descending in amplitude versus time, the method further comprising: Determining the transition point of the signal wave based at least in part on the end of the fall of the amplitude of the signal wave.   17. The method of claim 16, wherein the second time window is at least partially forward in time from the first time window, and the signal wave is rising in amplitude versus time, the method further comprising: Determining a transition point of the signal wave based at least in part on the end of the increase in amplitude of the signal wave.   The method according to claim 17, wherein the transition point of the signal wave is the start of the P wave of the electrocardiogram, the end of the P wave of the electrocardiogram, the end of the PR portion of the electrocardiogram, the point Q of the electrocardiogram, A method selected from the group consisting of an S point of an electrocardiogram, a start of a T wave of the electrocardiogram, an end of the T wave of the electrocardiogram, a start of a U wave of the electrocardiogram, and an end of the U wave of the electrocardiogram .   The method according to claim 18, wherein the transition point of the signal wave is the start of the P wave of the electrocardiogram, the end of the PR portion of the electrocardiogram, the point R of the electrocardiogram, the point J of the electrocardiogram, The method is selected from the group consisting of the start of a T wave, the start of a U wave of the electrocardiogram, and the end of a U wave of the electrocardiogram.   17. The method of claim 16, wherein the second time window is at least partially retrograde in time from the first time window, and the signal wave is decreasing in amplitude versus retrograde time, the method further comprising: Determining a transition point of the signal wave based at least in part on the end of the fall of the amplitude of the signal wave at a retrograde time.   23. The method according to claim 21, wherein the transition point of the signal wave is the start of the P wave of the electrocardiogram, the end of the P wave of the electrocardiogram, the point Q of the electrocardiogram, the point S of the electrocardiogram, and the point J of the electrocardiogram. , Selected from the group consisting of the end of the ECG T-wave, the start of the ECG U-wave, and the end of the ECG U-wave.   17. The method of claim 16, wherein the second time window reverses time at least partially from the first time window, and the signal wave rises in amplitude versus retrograde time, the method further comprising: Determining a transition point of the signal wave based at least in part on the end of the increase in amplitude of the signal wave at a retrograde time.   24. The method according to claim 23, wherein the transition point of the signal wave is the end of the P wave of the electrocardiogram, the end of the PR portion of the electrocardiogram, the point R of the electrocardiogram, the point J of the electrocardiogram, A method selected from the group consisting of the start of a T wave, the end of a T wave of the electrocardiogram, the start of a U wave of the electrocardiogram, and the end of a U wave of the electrocardiogram.   The method according to claim 3, further comprising comparing a transition point of each signal wave of the normal population of interest with the transition point of the determined signal wave to determine a relative position of the transition point of the determined signal wave. Determining whether the application of therapy has affected.   26. The method of claim 25, further comprising changing or stopping application of the treatment based at least in part on the comparison of the signal wave transition points.   27. The method of claim 26, wherein the treatment comprises chemotherapy.   27. The method according to claim 26, wherein the determined transition point of the signal wave is the end point of the T wave at which the electrocardiogram signal falls. In a system for determining the transition point of a signal wave,
a. A memory element configured to store data related to one or more signal waves sampled from an electrode operatively connected to one or more tissue structures;
b. A processor operatively connected to the memory element and configured to access data, the processor determining a transition point associated with the one or more signal waves;
The processor is
1) Sampling a first plurality of points of a signal wave of a first time window, wherein the first plurality of points at least a first time point and a last time point of the first time window Including
2) Sampling a second plurality of points of the signal wave in a second time window that is different in time from the first time window, wherein the second plurality of points is the first of the second time window Including at least a time point and a last time point,
3) compare the values of the first plurality of points with each other to determine if the in-window pattern rule is violated;
4) A secondary analysis is performed after determining that the in-window pattern rule has been violated, and the secondary analysis compares the second plurality of values with each other to determine whether the in-window pattern rule has been violated. A system characterized by being configured to determine.   30. The system of claim 29 further comprising a display operably connected to the processor, the processor further configured to cause the display to display at least one graphic image of the one or more signal waves. And a graphical representation of transition points associated with each of the one or more signal waves.   31. The system of claim 29 or 30, wherein the processor and memory element are operatively connected to one or more printed circuit boards comprising a card having a common housing and an electronic interface bus, the card being a personal A system that is configured to work with a computer.   32. The system of claim 29 or 30, wherein the processor and memory element comprise an application specific integrated circuit or field programmable gate array configured to be incorporated into a parent electronic device.   31. A system according to claim 29 or 30, wherein the processor and memory element comprise a personal computer.   34. The operating room electronic device of claim 29, wherein the processor and memory element are selected from the group consisting of an electrophysiological mapping system, an echocardiography system, and a fluoroscopic imaging system. A system characterized in that it is operatively connected to.   The system according to any one of claims 29 to 34, wherein the processor and the memory element are selected from the group consisting of an electrocardiogram system, an electroencephalogram system, an electromyogram system, and an emergency holter monitor. A system characterized in that it is operatively connected to.   31. The system of claim 29 or 30, wherein the processor and memory element are contained in a housing that can be incorporated, and the memory element is configured to exchange data with an external computer system via wireless transmission. A system characterized by   37. A system as claimed in any one of claims 29 to 36, wherein the processor samples a second plurality of points in a second time window forward at least partially in time from the first time window. A system characterized by being configured.   38. The system of claim 37, wherein the signal wave including the second plurality of sampled points is descending in amplitude versus time, and the transition point of the signal wave is based on the end of the descending of the descending signal wave. Determined by the processor.   38. The system of claim 37, wherein the signal wave including the second plurality of sampled points rises in amplitude versus time, and a signal wave transition point is based on the end of the rising signal wave rise. Determined by the processor.   37. A system as claimed in any one of claims 29 to 36, wherein the processor samples a second plurality of points in a second time window that is at least partially in time from the first time window. A system characterized by being configured.   41. The system of claim 40, wherein the signal wave including the second plurality of points to be sampled falls in amplitude versus retrograde time, and the transition point of the signal wave falls in the descending signal wave in retrograde time. Determined by the processor based on the end of the system.   41. The system of claim 40, wherein the signal wave including the second plurality of sampled points rises in amplitude versus retrograde time, and the transition point of the signal wave rises in the signal wave that rises in retrograde time. Determined by the processor based on the end of the system.

Images