WO2025006088A1 - Cardiac signal t-wave end time detection using maximum point of gradient signal - Google Patents

Abstract

An example device for detecting one or more parameters of a cardiac signal is disclosed herein. The device includes one or more electrodes and sensing circuitry configured to sense a cardiac signal via the one or more electrodes. The device further includes processing circuitry configured to determine a representative signal based on the cardiac signal, the representative signal having a single polarity, and determine a T-wave end time of the cardiac signal based on a maximum point on a gradient signal of the cardiac signal.

Description

CARDIAC SIGNAL T- WAVE END TIME DETECTION USING MAXIMUM POINT OF GRADIENT SIGNAL

[0001] This application claims the benefit of priority from U.S. Provisional Patent Application No. 63/511,396, filed June 30, 2023, the entire content of which is incorporated herein by reference.

TECHNICAL FIELD

[0002] This disclosure relates to cardiac monitoring and, more particularly, to evaluation of features of cardiac signals.

BACKGROUND

[0003] Cardiac signal analysis may be performed by a variety of devices, such as implantable medical devices (IMDs) and external devices (e.g., smart watches, fitness monitors, mobile devices, Holter monitors, wearable defibrillators, or the like). For example, devices may be configured to process cardiac signals (e.g., cardiac electrograms (EGMs) and electrocardiograms (ECGs)) sensed by one or more electrodes. Features of cardiac signals may include the P-wave, Q-wave, R-wave, S-wave, QRS-complex, and T-wave. Accurate detection and delineation of features cardiac signals, such as T-waves, may be of importance for improving operation of devices. As examples, detection of an occurrence of T-waves may be used to better identify clinically significant cardiac intervals, such as a QT interval or another cardiac activation recovery interval, or time delivery of therapy from the IMD.

[0004] Cardiac pacing is one type of therapy delivered to patients to treat a wide variety of cardiac dysfunctions. Cardiac pacing is often delivered by the IMD. An implantable cardioverter-defibrillator (ICD) may provide pacing functionality and also provide cardioversion or defibrillation (referred to as antitachyarrhythmia shock therapy) in response to a detection of a cardiac tachyarrhythmias, if needed. Detection of a variety of features of the cardiac signal may be used by the IMD to determine whether IMD should deliver therapy, for example pacing or antitachyarrhythmia shock therapy, to the patient. SUMMARY

[0005] In general, the disclosure is directed to devices and techniques for identifying one or more features and/or determining one or more parameters of a cardiac signal (e.g., EGM and/or ECG) of a patient. For example, the disclosure describes techniques for identifying the end of a T-wave, which may allow a more robust delineation of cardiac signal features. This improved identification of the end of the T-wave may, for example, allow determination of variability of the activation recovery interval (ARI), e.g., a QT interval or interval measured from the start of the QRS complex to the end of the T-wave, which may enable determining whether a patient is experiencing or will experience a tachyarrhythmia. In some examples, a determination that a patient is experiencing or will experience a tachyarrhythmia may enable an IMD to deliver therapy to the patient to terminate or prevent a predicted tachyarrhythmia, which therapy need not necessarily include an anti-tachyarrhythmia shock. Because, anti-tachyarrhythmia shocks may cause significant patient discomfort and do not always terminate a lethal tachyarrhythmia, the ability to deliver a non-shock therapy that prevents or terminates a predicted tachyarrhythmia may be considered a significant improvement to the operation of a medical device system to monitor and treat a patient.

[0006] Signal processing techniques that may be used by processing circuitry, e.g., of the IMD, to delineate the one or more features may include, for example, determining a programmable energy point or a maximum point of a gradient signal. The gradient signal may be calculated from a representative signal created based on the cardiac signal by derivation, and has a single polarity. The processing circuitry may be configured to determine the end of a T-wave based on determining a programmable energy point or a maximum point of a gradient signal of the cardiac signal. The processing circuitry may blank a portion of the cardiac signal to prevent some features of the signal (e.g., QRS complex) from interfering with the processing circuitry determining a programmable energy point or maximum of the gradient signal.

[0007] The processing circuitry may determine, based on a portion of the signal that has not been blanked, a squared gradient signal having a programmable energy point. By multiplying the gradient signal by an exponential (e.g., squaring the blanked signal), the processing circuitry may determine a squared gradient signal. Using a technique to determine the area under the curve, the IMD may set, as a predetermined percentage value, a programmable energy point corresponding to a quotient based on an area under the squared gradient curve. The quotient may result from the division of a portion of the area under the squared gradient signal divided by the total area under the squared gradient signal. The processing circuitry may determine the programmable energy point by calculating a plurality of quotients based on a plurality of area portions under the squared gradient signal. If a quotient from the plurality of quotients is equal to the predetermined percentage value, the processing circuitry may determine a point, corresponding to the quotient, is the programmable energy point. The programmable energy point may be determined as a point at which a predetermined percentage of the area under the T-wave curve is earlier in time than the programmable energy point. For example, the programmable energy point may be set such that 60% of the area under the entire T-wave exists earlier than the programmable energy point. For example, the programmable energy point may correspond with a time value by which the predetermined percentage of energy in the T-wave signal has been expended. In some examples, the predetermined percentage may be 60%. For additional accuracy, the time value may be used to center a T-wave window on the cardiac signal and take the gradient of the cardiac signal within the window. A maximum point of the gradient signal may be calculated to determine an occurrence of a T-wave end time.

[0008] In one example, a technique may include determining, by processing circuitry of a medical device system and based on a cardiac signal sensed by a medical device of the medical device system, a digital signal comprising a plurality of digital samples representing the cardiac signal. The technique may also include determining, by the processing circuitry and based on a time window of the plurality of digital samples, a gradient signal of a set of digital samples of the plurality determining, by the processing circuitry and during the time window, a maximum point of the gradient signal. The technique may include storing on a memory device, by the processing circuitry and based on the maximum point, data indicating an occurrence of a T-wave at a corresponding time value.

[0009] In one example, a medical device may include a memory device, sensing circuitry, and processing circuitry. The sensing circuitry of the medical device may be configured to sense a cardiac signal via a plurality of electrodes. The processing circuitry of the medical device may be configured to determine, by the processing circuitry based on the cardiac signal sensed by the sensing circuitry of the medical device, a digital signal comprising a plurality of digital samples representing the cardiac signal. The processing circuitry may be further configured to determine, by the processing circuitry and based a time window of the plurality of digital samples, a gradient signal of a set of digital samples. The processing circuitry may be further configured to determine, by the processing circuitry and during the time window, a maximum point of the gradient signal; and Processing circuitry may additionally be configured to store on the memory device, by the processing circuitry and based on the maximum point, data indicating an occurrence of a T-wave at a corresponding time value.

[0010] In another example, a system includes an IMD configured to sense a cardiac signal. The system may include processing circuitry configured to determine, by the processing circuitry and based on the cardiac signal sensed by the implantable medical device of the system, a digital signal comprising a plurality of digital samples representing the cardiac signal. The processing circuitry may be further configured to determine, by the processing circuitry and based on a time window of the plurality of digital samples, a gradient signal of a set of digital samples. The processing circuitry may be configured to determine, by the processing circuitry and during the time window, a maximum point of the gradient signal. The processing circuitry may further be configured to store on a memory device of the implantable medical device, by the processing circuitry and based on the maximum point, data indicating an occurrence of a T-wave at a corresponding time value. [0011] The summary is intended to provide an overview of the subject matter described in this disclosure. It is not intended to provide an exclusive or exhaustive explanation of the systems, device, and methods described in detail within the accompanying drawings and description below. Further details of one or more examples of this disclosure are set forth in the accompanying drawings and in the description below. Other features, objects, and advantages will be apparent from the description and drawings, and from the claims.

BRIEF DESCRIPTION OF DRAWINGS

[0012] FIG. 1 is a conceptual diagram illustrating an example system for monitoring and treating cardiac events, which may be used to identify one or more parameters of a cardiac electrogram (EGM). [0013] FIG. 2 is a conceptual diagram illustrating the IMD and leads of the system of FIG. 1 in greater detail.

[0014] FIG. 3 is a block diagram illustrating an example configuration of an IMD for monitoring and treating cardiac events, which may be used to identify one or more parameters of the cardiac EGM.

[0015] FIG. 4 is a flow diagram illustrating a first example method of determining a parameter of the cardiac EGM according to the techniques of this disclosure.

[0016] FIG. 5 is a voltage/time graph illustrating a plot of an example of an EGM signal plotted over time, in accordance with one or more techniques of this disclosure.

[0017] FIG. 6 is a voltage/time graph illustrating an example of an EGM signal before and after applying a blanking and gradient techniques, according to the techniques of this disclosure.

[0018] FIG. 7 is a voltage/time graph illustrating an example of a squared gradient signal calculated based on a gradient signal, according to the techniques of this disclosure.

[0019] FIG. 8 is a zoomed in voltage/time graph illustrating an example comparison between a first time value determined based on a programmable energy point and a second time value determined based on a maximum point, according to the techniques of this disclosure.

[0020] FIG. 9 is a flow diagram illustrating an example technique for determining a time value corresponding to an occurrence of a T-wave end time, according to the techniques of this disclosure.

[0021] FIG. 10 is a flow diagram illustrating an example technique for determining a time value using a maximum technique based on a filtered ECG signal, according to the techniques of this disclosure.

DETAILED DESCRIPTION

[0022] This disclosure describes techniques for identifying one or more parameters of a cardiac signal, such as end times of the QRS complex and T-wave. The parameters may be used to, for example, detect or predict arrhythmias, to evaluate other conditions of the patient, or to configure and/or evaluate therapies, such as CRT. [0023] In some examples, IMDs may deliver therapy to a patient based on observable events (e.g., a tachyarrhythmia) of a sensed cardiac EGM. Although some therapies such as defibrillation shocks may restore the patient’s heart to normal function after a tachyarrhythmia is detected, defibrillation shocks may be painful to the patient and cause permanent damage to the patient’s cardiac tissue. In some cases, the patient may require hospitalization following the delivery of a defibrillation shock. For at least these reasons, it may be beneficial to improve an accuracy in which an IMD detects the presence of an arrhythmia in the patient, e.g., by increasing an accuracy of the one or more parameters of the cardiac EGM measured by the IMD. Furthermore, it may be beneficial to predict tachyarrhythmia before it occurs and deliver therapy (such as pacing) to prevent the tachyarrhythmia, rather than shocks to terminate an occurring tachyarrhythmia.

[0024] Techniques of this disclosure may increase the accuracy with which processing circuitry identifies one or more parameters of a cardiac signal acquired by a device, such as, for example, start and/or end times of one or more features of a cardiac signal (e.g., P-wave start time, P-wave end time, QRS complex start time, QRS complex end time, T-wave start time, and T-wave end time), which may allow for more accurate delineation of these features. In some examples, the processing circuitry may determine representative signals of the cardiac signal by performing signal processing methods including but not limited to filtering the cardiac signal, calculating the gradient of the cardiac signal, and amplifying the cardiac signal. The processing circuitry may determine an end time of a QRS complex of the cardiac signal and an end time of the subsequent T-wave in the cardiac signal based on areas under the representative signals calculated by the device. T-wave end times determined according to the techniques described herein may be used, for example, to calculate shortterm variability (STV) in the activation recovery interval (ARI) and/or QT interval, which may in turn be used to predict tachyarrhythmia and or determine whether CRT will be effective, as examples.

[0025] In some examples, the techniques of this disclosure may enable identification of end times that accurately reflect times in which the corresponding events occurred in the heart by processing circuitry using a cardiac EGM acquired by an IMD, rather than an electrocardiogram (ECG). Furthermore, the computational complexity of the techniques described herein may be appropriate for implementation by processing circuitry of an IMD. In other examples, the techniques of this disclosure may enable accurate identification of features of an ECG acquired by an external device, such as but not limited to a smart watch, a fitness monitor, a mobile device, a Holter monitor, or a wearable defibrillator.

[0026] FIG. 1 is a conceptual diagram illustrating an example system 10 for monitoring and treating cardiac events, which may be used to identify one or more parameters of a cardiac electrogram (EGM). As illustrated by example system 10 in FIG. 1, a system for identifying one or more parameters of the cardiac EGM according to the techniques of this disclosure may include an IMD 16, which in the illustrated example is an ICD with pacing capabilities. IMD 16 is connected to leads 18, 20 and 22 and is communicatively coupled to external device 24. IMD 16 senses electrical signals attendant to the depolarization and repolarization of heart 12, e.g., an EGM, via electrodes on one or more leads 18, 20 and 22 or the housing of IMD 16. IMD 16 may also deliver therapy in the form of electrical signals to heart 12 via electrodes located on one or more leads 18, 20 and 22 or a housing of IMD 16. The therapy may be pacing, cardioversion and/or defibrillation pulses. IMD 16 may monitor EGM signals collected by electrodes on leads 18, 20 or 22, and based on the EGM signal, diagnose and treat cardiac episodes.

[0027] Leads 18, 20, 22 extend into the heart 12 of patient 14 to sense electrical activity of heart 12 and/or deliver electrical stimulation to heart 12. In the example shown in FIG. 1, right ventricular (RV) lead 18 extends through one or more veins (not shown), the superior vena cava (not shown), and right atrium 26, and into right ventricle 28. Left ventricular (LV) lead 20 extends through one or more veins, the vena cava, right atrium 26, and into the coronary sinus 30 to a region adjacent to the free wall of left ventricle 32 of heart 12. Right atrial (RA) lead 22 extends through one or more veins and the vena cava, and into the right atrium 26 of heart 12.

[0028] In some examples, external device 24 takes the form of a handheld computing device, computer workstation or networked computing device that includes a user interface for presenting information to and receiving input from a user. A user, such as a physician, technician, surgeon, electro-physiologist, or other clinician, may interact with external device 24 to retrieve physiological or diagnostic information from IMD 16. A user may also interact with external device 24 to program IMD 16, e.g., select values for operational parameters of the IMD. External device 24 may include processing circuitry configured to evaluate EGM signals transmitted from IMD 16 to external device 24.

[0029] IMD 16 and external device 24 may communicate via wireless communication using any techniques known in the art. Examples of communication techniques may include, for example, low frequency or radiofrequency (RF) telemetry, or according to the Bluetooth® or Bluetooth LE specifications. In some examples, external device 24 may be located remotely from IMD 16 and communicate with IMD 16 via a network.

[0030] System 10 of FIG. 1 is an example of a system for identifying one or more parameters of a cardiac EGM according to the techniques of this disclosure. In some examples, processing circuitry of one or both of IMD 16 and external device 24 includes cardiac signal analysis circuitry configured to determine one or more parameters of a cardiac signal of patient 14. In one example, the cardiac signal includes a cardiac EGM sensed via one or more electrodes of IMD 16. A cardiac EGM is a signal representative of electrical activity of the heart, measured by electrodes implanted within the body, and often within the heart itself. For example, a cardiac EGM may include P-waves (depolarization of the atria), R-waves (depolarization of the ventricles), and T-waves (repolarization of the ventricles), among other events. Information relating to the aforementioned events, such as time separating one or more of the events, may be applied for a number of purposes, such as to determine whether an arrhythmia is occurring and/or predict whether an arrhythmia is likely to occur. Cardiac signal analysis circuitry, which may be implemented as part of processing circuitry of IMD 16 and/or external device 24, may perform signal processing techniques to extract information indicating the one or more parameters of the cardiac signal.

[0031] IMD 16 may be configured to determine a time representing a start of a QRS complex of the cardiac EGM, a time representing an end of the QRS complex, and a time representing an end of a subsequent T-wave. In one example, IMD 16 is further configured to determine the QT interval, or activation recovery interval (ARI), as a time separating the start of the QRS complex and the end of the subsequent T-wave. Subsequently, IMD 16 may measure a plurality of QT intervals, each QT interval of the plurality of QT intervals corresponding to a heartbeat of a plurality of heartbeats. The plurality of QT intervals, and other parameters such as but not limited to T-wave duration and QRS duration, may predict an upcoming (e.g., occurring within about one minute to about five minutes from the prediction) arrhythmia in patient 14. In response to determining that an arrhythmia will occur in patient 14, IMD 16 may deliver therapy to heart 12 of patient 14 via one or more electrodes of leads 18, 20, and 22.

[0032] Although the techniques for identifying one or more parameters of the cardiac EGM according to the techniques of this disclosure are described herein primarily with reference to example system 10, the techniques may be performed by other systems that differ from example system 10. For example, systems for identifying the one or more parameters according to the techniques of this disclosure may include an IMD having different functionality than IMD 16, and may include more, fewer, or different implantable cardiac leads than leads 18, 20 and 22. In some examples, systems for identifying the one or more parameters include more or fewer leads, do not include any intracardiac leads, or do not include any leads. Example IMDs that may implement the techniques of this disclosure in addition to the illustrated example of IMD 16 include extracardiovascular ICDs, transcatheter pacing systems, such as the Micra™ transcatheter pacing system commercially available from Medtronic, Inc., of Minneapolis, Minnesota, and insertable cardiac monitors, such as the Reveal LINQ™ or LINQ II™ available from Medtronic, Inc.

[0033] FIG. 2 is a conceptual diagram illustrating IMD 16 and leads 18, 20 and 22 of system 10 in greater detail. In the illustrated example, bipolar electrodes 40 and 42 are located adjacent to a distal end of lead 18, and bipolar electrodes 48 and 50 are located adjacent to a distal end of lead 22. In addition, four electrodes 44, 45, 46 and 47 are located adjacent to a distal end of lead 20. Lead 20 may be referred to as a quadripolar LV lead. In other examples, lead 20 may include more or fewer electrodes. In some examples, LV lead 20 includes segmented electrodes, e.g., in which each of a plurality of longitudinal electrode positions of the lead, such as the positions of electrodes 44, 45, 46 and 47, includes a plurality of discrete electrodes arranged at respective circumferential positions around the circumference of lead.

[0034] In the illustrated example, electrodes 40, 44-47 and 48 take the form of ring electrodes, and electrodes 42 and 50 may take the form of extendable helix tip electrodes mounted retractably within insulative electrode heads 52 and 56, respectively. Leads 18 and 22 also include elongated electrodes 62 and 66, respectively, which may take the form of a coil. In some examples, each of electrodes 40, 42, 44-48, 50, 62, and 66 is electrically coupled to a respective conductor within the lead body of its associated lead 18, 20, 22 and thereby coupled to circuitry within IMD 16.

[0035] In some examples, IMD 16 includes one or more housing electrodes, such as housing electrode 4 illustrated in FIG. 2, which may be formed integrally with an outer surface of hermetically-sealed housing 8 of IMD 16 or otherwise coupled to housing 8. In some examples, housing electrode 4 is defined by an uninsulated portion of an outward facing portion of housing 8 of IMD 16. Other divisions between insulated and uninsulated portions of housing 8 may be employed to define two or more housing electrodes. In some examples, a housing electrode includes substantially all of housing 8.

[0036] Housing 8 encloses signal generation circuitry that generates therapeutic stimulation, such as cardiac pacing, cardioversion and defibrillation pulses, as well as sensing circuitry for sensing electrical signals attendant to the depolarization and repolarization of heart 12. Housing 8 may also enclose a memory for storing the sensed electrical signals. Housing 8 may also enclose telemetry circuitry for communication between IMD 16 and external device 24.

[0037] IMD 16 senses electrical signals attendant to the depolarization and repolarization of heart 12 via electrodes 4, 40, 42, 44-48, 50, 62, and 66. IMD 16 may sense such electrical signals via any bipolar combination of electrodes 40, 42, 44-48, 50, 62, and 66. Furthermore, any of the electrodes 40, 42, 44-48, 50, 62, and 66 may be used for unipolar sensing in combination with housing electrode 4.

[0038] The illustrated numbers and configurations of leads 18, 20 and 22 and electrodes are merely examples. Other configurations, i.e., number and position of leads and electrodes, are possible. In some examples, system 10 may include an additional lead or lead segment having one or more electrodes positioned at different locations in the cardiovascular system for sensing and/or delivering therapy to patient 14. For example, instead of or in addition to intracardiac leads 18, 20 and 22, system 10 may include one or more extracardiovascular (e.g., epicardial, substemal, or subcutaneous) leads not positioned within the heart.

[0039] FIG. 3 is a block diagram illustrating an example configuration of IMD 16 for monitoring and treating cardiac events, which may be used to identify one or more parameters of a cardiac EGM according to the techniques of this disclosure. In the illustrated example, IMD 16 includes processing circuitry 70, memory 72, signal generation circuitry 74, sensing circuitry 76, telemetry circuitry 78, cardiac signal analysis circuitry 80, and activity sensor 82. Memory 72 includes computer-readable instructions that, when executed by processing circuitry 70, cause IMD 16 and processing circuitry 70 to perform various functions attributed to IMD 16 and processing circuitry 70 herein. Memory 72 may include any volatile, non-volatile, magnetic, optical, or electrical media, such as a random access memory (RAM), read-only memory (ROM), non-volatile RAM (NVRAM), electrically- erasable programmable ROM (EEPROM), flash memory, or any other digital or analog media.

[0040] Processing circuitry 70 may include any one or more of a microprocessor, a controller, a digital signal processor (DSP), a graphics processing unit (GPU), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or equivalent discrete or analog logic circuitry. In some examples, processing circuitry 70 may include multiple components, such as any combination of one or more microprocessors, one or more controllers, one or more DSPs, one or more ASICs, or one or more FPGAs, as well as other discrete or integrated logic circuitry. The functions attributed to processing circuitry 70 herein may be embodied as software, firmware, hardware or any combination thereof.

[0041] Generally, processing circuitry 70 controls signal generation circuitry 74 to deliver therapy to heart 12 of patient 14 according to selected values of one or more of therapy parameters, which may be stored in memory 72. As an example, processing circuitry 70 may control signal generation circuitry 74 to deliver electrical pulses with the amplitudes, pulse widths, frequency, and/or electrode polarities specified by the selected therapy parameter values. Parameters stored in memory 72 include thresholds or other conditions, which may be compared to parameters of the EGM, and based on which processing circuitry 70 controls signal generation circuitry 74 to deliver therapy, such as cardiac rates, intervals, and/or EGM morphology parameters.

[0042] Signal generation circuitry 74 is configured to generate and deliver electrical therapy to patient 14. As shown in FIG. 3, signal generation circuitry 74 is electrically coupled to electrodes 4, 40, 42, 44-48, 50, 62, and 66, e.g., via conductors of the respective leads 18, 20, and 22 and, in the case of housing electrode 4, within housing 8. For example, signal generation circuitry 74 may deliver pacing, defibrillation or cardioversion pulses to heart 12 via at least two of electrodes 4, 40, 42, 44-48, 50, 62 and 66. In some examples, signal generation circuitry 74 delivers therapy in the form of signals other than pulses such as sine waves, square waves, or other substantially continuous time signals.

[0043] Signal generation circuitry 74 may include one or more capacitors, charge pumps, current sources, or other signal generation circuitry. Signal generation circuitry 74 may also include switching circuitry (not shown) and processing circuitry 70 may use the switching circuitry to select, e.g., via a data/address bus, which of the available electrodes are used to deliver the electrical stimulation. The switching circuitry may include a switch array, switch matrix, multiplexer, or any other type of switching device suitable to selectively couple energy to selected electrodes.

[0044] Electrical sensing circuitry 76 monitors electrical cardiac signals from any combination of electrodes 4, 40, 42, 44-48, 50, 62 and 66. Sensing circuitry 76 may also include switching circuitry which processing circuitry 70 controls to select which of the available electrodes are used to sense the heart activity, depending upon which electrode combination is used in the current sensing configuration. In some examples, sensing circuitry 76 may include one or more amplifiers, filters, and analog-to-digital converters.

[0045] Sensing circuitry 76 may include one or more detection channels, each of which may include an amplifier. The detection channels may be used to sense cardiac signals, such as a cardiac EGM. Some detection channels may detect events, such as R-waves, P-waves, and T-waves and provide indications of the occurrences of such events to processing circuitry 70. One or more other detection channels may provide the signals to an analog-to-digital converter, for conversion into a digital signal for processing or analysis by processing circuitry 70 or cardiac signal analysis circuitry 80.

[0046] For example, sensing circuitry 76 may include one or more narrow band channels, each of which may include a narrow band filtered sense-amplifier that compares the detected signal to a threshold. If the filtered and amplified signal is greater than the threshold, the narrow band channel indicates that a certain electrical cardiac event, e.g., depolarization, has occurred. Processing circuitry 70 then uses that detection in measuring frequencies of the sensed events. In one example, at least one narrow band channel may include an R-wave or P-wave amplifier.

[0047] In some examples, sensing circuitry 76 includes a wide band channel which may include an amplifier with a relatively wider pass band than the narrow band channels. Signals from the electrodes that are selected for coupling to the wide-band amplifier may be converted to multi-bit digital signals by an analog-to-digital converter (ADC) provided by, for example, sensing circuitry 76 or processing circuitry 70. Processing circuitry 70 and cardiac signal analysis circuitry 80 may analyze the digitized version of signals from the wide band channel. Processing circuitry 70 may employ digital signal analysis techniques to characterize the digitized signals from the wide band channel to, for example, detect and classify the patient’s heart rhythm.

[0048] Processing circuitry 70 may detect and classify the patient’s heart rhythm based on the cardiac electrical signals sensed by sensing circuitry 76 employing any of the numerous signal processing methodologies known in the art. For example, processing circuitry 70 may maintain escape interval counters that may be reset upon sensing of R- waves by sensing circuitry 76. The value of the count present in the escape interval counters when reset by sensed depolarizations may be used by an episode classifier of processing circuitry 70 to measure the durations of R-R intervals, which are measurements that may be stored in memory 72. Processing circuitry 70 may use the count in the interval counters to detect a tachyarrhythmia, such as ventricular fibrillation or ventricular tachycardia. A portion of memory 72 may be configured as a plurality of recirculating buffers, capable of holding series of measured intervals, which may be analyzed by processing circuitry 70 to determine whether the patient's heart 12 is presently exhibiting atrial or ventricular tachyarrhythmia. [0049] In some examples, processing circuitry 70 may determine that tachyarrhythmia has occurred by identification of shortened R-R interval lengths. For example, processing circuitry 70 detects tachycardia when the interval length falls below a first threshold (e.g., 360 milliseconds (ms)) and fibrillation when the interval length falls below a second threshold (e.g., 320 ms). These interval lengths are merely examples, and a user may define the interval lengths as desired, which may then be stored within memory 72. This interval length may need to be detected for a certain number of consecutive cycles, for a certain percentage of cycles within a running window, or a running average for a certain number of cardiac cycles, as examples. In some examples, an arrhythmia detection method may include any combination of suitable tachyarrhythmia detection algorithms. For example, EGM morphology may be considered in addition to or instead of interval length for detecting or predicting tachyarrhythmias. [0050] Generally, processing circuitry 70 detects a treatable tachyarrhythmia, such as VF, based on the EGM, e.g., the R-R intervals and/or morphology of the EGM, and selects a therapy to deliver to terminate the tachyarrhythmia, such as an anti-tachycardia pacing regimen and/or defibrillation pulse of a specified magnitude. The detection of the tachyarrhythmia may include a number of phases or steps prior to delivery of the therapy, such as first phase, sometimes referred to as detection, in which a number of consecutive or proximate R-R intervals satisfies a first number of intervals to detect (NID) criterion, a second phase, sometimes referred to as confirmation, in which a number of consecutive or proximate R-R intervals satisfies a second, more restrictive NID criterion. Tachyarrhythmia detection may also include confirmation based on EGM morphology or other sensors subsequent to or during the second phase.

[0051] One or more sensors 82 may be optionally included in some examples of IMD 16. Sensors 82 may include one or more accelerometers. Sensors 82 may additionally or alternatively include other sensors such as a heart sounds sensor, a pressure sensor, a flow sensor, or an oxygen (O2) saturation sensor. In some examples, processing circuitry 70 may detect respiration via a signal received from sensing circuitry 76 via one or more electrodes. [0052] Information obtained from activity sensor 82 may be used to determine activity level, posture, blood pressure, blood flow, blood oxygen level, or respiratory rate, as examples. In some examples, this information may be used by IMD 16 to aid in the classification of an abnormal heart rhythm. In some examples, this information may be used by IMD 16 or a user of external device 24 to determine desired LV pacing locations and timings for delivery of CRT. For example, blood pressure or flow metrics may indicate the effectiveness LV pacing locations and timings in improving the performance of heart 12.

[0053] Sensors 82 may be located outside of the housing 8 of IMD 16. Sensors 82 may be located on a lead that is coupled to IMD 16 or may be implemented in a remote sensor that wirelessly communicates with IMD 16 via telemetry circuitry 78. In any case, sensors 82 are electrically or wirelessly coupled to circuitry contained within housing 8 of IMD 16. [0054] In some examples, IMD 16 includes cardiac signal analysis circuitry 80. Cardiac signal analysis circuitry 80 and IMD 16 are configured to perform techniques for identifying one or more parameters of the cardiac EGM, as described herein. Cardiac signal analysis circuitry 80 may include software and/or firmware executed by processing circuitry 70. Additionally, or alternatively, cardiac signal analysis circuitry 80 may comprise certain circuitry of processing circuitry 70. According to some examples, sensing circuitry 76 senses a cardiac signal (e.g., cardiac EGM) via any combination of electrodes 4, 40, 42, 44-48, 50, 62, and 66 of IMD 16. As discussed above, sensing circuitry 76 may include a wide-band amplifier, and sensing circuitry 76 may sense the cardiac signal with the wide-band sensing amplifier. In response to detecting or predicting an arrhythmia based on the one or more parameters identified by cardiac signal analysis circuitry 80, IMD 16 may deliver therapy pulses to patient 14. In one example, under the control of processing circuitry 70, signal generation circuitry 74 delivers therapy pulses to heart 12 via one or more of the electrodes, e.g., 40, 42, 44-48, 50, 62, and 66, of leads 18, 20, and 22.

[0055] Cardiac signal analysis circuitry 80 may include a microprocessor, a microcontroller, a DSP, a GPU, an ASIC, an FPGA, or other equivalent discrete or integrated logic circuitry. Accordingly, cardiac signal analysis circuitry 80 may include any suitable structure, whether in hardware, software, firmware or any combination thereof, to perform the functions ascribed herein to cardiac signal analysis circuitry 80. In one example, cardiac signal analysis circuitry 80 receives a cardiac EGM via sensing circuitry 76, e.g., the wide band sensing channel of sensing circuitry 76, and any combination of electrodes 4, 40, 42, 44-48, 50, 62, and 66. Based on representative signals of the cardiac EGM, cardiac signal analysis circuitry 80 may measure a plurality of QT intervals of the cardiac EGM, each QT interval of the plurality of QT intervals representing a window of time between a start time of a QRS complex of the cardiac EGM, and an end time of a T-wave of the cardiac EGM (i.e., a window of time between the beginning of ventricular depolarization and the ending of ventricular repolarization). Additionally, or alternatively, cardiac signal analysis circuitry 80 may measure other parameters such as T-wave duration (window of time separating T-wave start time and T-wave end time), QRS duration, and heart rate.

[0056] Parameters (e.g., QRS end time, T-wave end time, QRS duration, T-wave duration, QT interval length, or the like) of the cardiac EGM measured by cardiac signal analysis circuitry 80 may indicate that an arrhythmia is imminent in patient 14. Hence, IMD 16 may perform analysis on parameters of the cardiac EGM to predict potential upcoming arrhythmia and deliver therapy to prevent the potential upcoming arrhythmia. In one example, IMD 16 may be configured to predict an upcoming arrhythmia between about one minute and about five minutes before an onset of the upcoming arrhythmia. After predicting the arrhythmia, IMD 16 may deliver therapy to heart 12 via signal generation circuitry 74. IMD 16 may deliver therapy via any combination of electrodes 40, 42, 44-48, 50, 62, and 66. IMD 16, as illustrated in FIG. 3, may accurately measure parameters of the cardiac EGM by producing determinations of QRS complex start times, QRS complex end times, and T-wave end times, among other parameters, that may be more accurate than determinations made by other techniques.

[0057] For example, cardiac signal analysis circuitry 80 may use a band-pass filter to create a filtered version of the EGM signal received from sensing circuitry 76. In one example, the band-pass filter may have a frequency pass range of 1 Hertz to 10 Hertz (e.g., the bandpass filter attenuates frequency bands outside the frequency pass range), thus eliminating high-frequency and low-frequency noise in the cardiac EGM signal. In one example, noise may be introduced in the cardiac EGM signal by the respiratory system, the muscular system, the digestive system, the nervous system, and other systems in patient 14, as well as 60 Hertz noise or other external noise at any frequency from external sources such as medical equipment or other machinery. Cardiac signal analysis circuitry 80, or other components of IMD 16, may detect an occurrence of an R-wave (e.g., ventricular depolarization event) in the EGM. For example, sensing circuitry 76, e.g., a narrow band R- wave detection channel within the sensing circuitry, may provide an indication of the timing of R-wave detection to cardiac signal analysis circuitry 80. Cardiac signal analysis circuitry 80 may set a QRS window based on the detected R-wave, the QRS window starting at a first time relative to the R-wave and ending at a second time relative to the R-wave, the first time and the second time of the QRS window being QRS window parameters 86 stored in memory 72 of IMD 16. In some examples, the first time and the second time of the QRS window both occur after the R-wave. In other examples, at least one of the first time and the second time may occur before the R-wave.

[0058] In one example, the first time occurs 10 milliseconds after the R-wave, and the second time occurs 150 milliseconds after the R-wave. In other examples, the first time and the second time of the QRS window may include other values. IMD 16 may update QRS window parameters 86 in response to any of a variety of events. In some examples, IMD 16 may receive instructions from external device 24 to update QRS window parameters 86 (e.g., change the first time and the second time of the QRS window). Additionally, or alternatively, processing circuitry 70 of IMD 16 may automatically update QRS window parameters 86 in response to detecting a change in a heart rate of patient 14. For example, IMD 16 may decrease a length of the QRS window if the heart rate of patient 14 increases; and IMD 16 may increase the length of the QRS window if the heart rate of patient 14 decreases. In other examples, IMD 16 may update QRS window parameters 86 in response to other events, such as but not limited to signals from sensor(s) 82.

[0059] The QRS window may represent a period of time in which an end of the QRS complex is expected to occur. Hence, when determining the end of the QRS complex, IMD 16 may eliminate (“blank”) portions of the filtered EGM outside of the QRS window. Additionally, IMD 16 may compute a gradient of the filtered EGM signal, the gradient being a derivative of the filtered EGM signal. A representative signal of the QRS complex may be calculated, the representative signal including the gradient of the filtered EGM signal. Additionally, or alternatively, the representative signal of the QRS complex may include an amplification of the gradient of the filtered EGM signal. In one example, the representative signal includes an exponential (e.g., square, cube, or the like) of the gradient of the filtered EGM signal. In one example, the representative signal may include a single polarity (e.g., the representative signal may be exclusively positive or exclusively negative, however, the representative signal may not alternate between positive and negative polarities). For instance, if the representative signal includes a square of the gradient of the filtered EGM signal, the square rectifies any negative portions of the representative signal, rendering the representative signal of the QRS complex to a single polarity. In other examples, IMD 16 may use a half-wave rectifier or a full-wave rectifier to create the representative signal having a single polarity. Additionally, or alternatively, IMD 16 may perform an absolute value function, a magnitude function, or the like to determine the representative signal having a single polarity.

[0060] Cardiac signal analysis circuitry 80 may determine an end of a QRS complex in the cardiac EGM signal based on an area under the representative signal of the QRS complex. In one example, cardiac signal analysis circuitry 80 is configured to compute a total area under the representative signal of the QRS complex. Additionally, cardiac signal analysis circuitry 80 is configured to calculate an area under any portion of the representative signal of the QRS complex, including any portion of the representative signal beginning at the first time of the QRS window and ending before the second time of the QRS window. Cardiac signal analysis circuitry 80 is configured to determine an end of a QRS complex as being an earliest determined time in which the area under the representative signal of the QRS complex is greater than or equal to a predetermined portion (e.g., threshold portion) of a total area under the representative signal of the QRS complex. In one example, the portion of the total area under the representative signal of the QRS complex is 0.9. In other words, cardiac signal analysis circuitry 80 determines the end of the QRS complex as being a first point within the QRS window in which the area under the representative signal of the QRS complex is greater than or equal to 90% of the total area under the representative signal. In other examples, the portion of the total area under the representative signal of the QRS complex indicating the end of the QRS complex may be greater or less than 0.9.

[0061] The portion of the total area under the representative signal of the QRS complex indicating the end of the QRS complex may be included within signal area threshold(s) 84 stored in memory 72 of IMD 16. In one example, IMD 16 may receive instructions from external device 24 to change signal area threshold(s) 84. In other examples, IMD 16 may update signal area threshold(s) 84 based on signals received from sensor(s) 82, processing circuitry 70, or cardiac signal analysis circuitry 80.

[0062] As discussed above, IMD 16 may detect an R-wave, and cardiac signal analysis circuitry 80 may set a QRS window based on the detected R-wave. Additionally, or alternatively, cardiac signal analysis circuitry 80 may set a T-wave window starting at a first time relative to the R-wave and ending at a second time relative to the R-wave, the first time and the second time of the T-wave window being T-wave window parameters 88 stored in memory 72 of IMD 16. In some examples, the first time of the T-wave window occurs at the QRS end time determined by cardiac signal analysis circuitry 80, and the second time of the T-wave window occurs 700 milliseconds after the R-wave. In other examples, the first time and the second time of the T-wave window may include other values. IMD 16 may update T- wave window parameters 88 based on any of a variety of events. In some examples, IMD 16 may receive instructions from external device 24 to update T-wave window parameters 88 (e.g., change the first time and the second time of the T-wave window). Additionally, or alternatively, processing circuitry 70 of IMD 16 may automatically update T-wave window parameters 88 in response to detecting a change in a heart rate of patient 14. For example, IMD 16 may decrease a length of the T-wave window if the heart rate of patient 14 increases; and IMD 16 may increase the length of the T-wave window if the heart rate of patient 14 decreases. In other examples, IMD 16 may update T-wave window parameters 88 in response to other events, such as but not limited to signals from sensor(s) 82.

[0063] The T-wave window may represent a period of time in which an end of the T- wave is expected to occur. Hence, when determining the end of the T-wave complex, IMD 16 may blank portions of the fdtered EGM outside of the T-wave window (e.g., blanked portions may include the QRS complex preceding the T-wave, and a P-wave following the T-wave). Additionally, IMD 16 may compute a gradient of the filtered EGM signal, the gradient being a derivative of the filtered EGM signal. A representative signal of the T-wave may be determined, the representative signal including the gradient of the filtered EGM signal. Additionally, or alternatively, the representative signal may include an amplification of the gradient of the filtered EGM signal. In some examples, the representative signal includes an exponential (e.g., square, cube, or the like) of the gradient of the filtered EGM signal. In some examples, the representative signal may include a single polarity (e.g., the representative signal may be exclusively positive or exclusively negative, however, the representative signal may not alternate between positive and negative polarities). For instance, if the representative signal includes a square of the gradient of the filtered EGM signal, the square rectifies any negative portions of the representative signal, rendering the representative signal of the T-wave to a single polarity.

[0064] Cardiac signal analysis circuitry 80 may determine an end of a T-wave in the cardiac EGM signal based on an area under the representative signal of the T-wave. In one example, cardiac signal analysis circuitry 80 is configured to compute a total area under the representative signal of the T-wave. Additionally, cardiac signal analysis circuitry 80 is configured to calculate an area under any portion of the representative signal of the T-wave, including any portion of the representative signal beginning at the first time of the T-wave window and ending before the second time of the T-wave window. Cardiac signal analysis circuitry 80 is configured to determine an end of a T-wave as being an earliest determined time in which the area under the representative signal of the T-wave is greater than or equal to a predetermined portion (e.g., threshold portion) of a total area under the representative signal of the T-wave. In one example, the predetermined portion of the total area under the representative signal of the T-wave is 0.6. In other words, cardiac signal analysis circuitry 80 determines the end of the T-wave as being a first point within the T-wave window in which the area under the representative signal of the T-wave is equal to 60% of the total area under the representative signal. In other examples, the portion of the total area under the representative signal indicating the end of the T-wave may be greater or less than 0.6.

[0065] The portion of the total area under the representative signal indicating the end of the T-wave may be included within signal area threshold(s) 84 stored in memory 72 of IMD 16. In one example, IMD 16 may receive instructions from external device 24 to change signal area threshold(s) 84. In other examples, IMD 16 may update signal area threshold(s) 84 based on signals received from sensor(s) 82, processing circuitry 70, or cardiac signal analysis circuitry 80.

[0066] In some examples, IMD may use a maximum gradient detection technique to determine an end of a T-wave. A cardiac signal detected by sensing circuitry 76 may be digitized by an ADC of sensing circuitry 76. For example, an ADC determine a digital signal including a plurality of digital samples of the cardiac signal. A bandpass filter, for example a digital filter of cardiac signal analysis circuitry 80, may filter the gradient signal into a filtered signal. For example, cardiac signal analysis circuity 80 may filter out noise from the plurality of digital samples resulting in a filtered EGM signal. Cardiac signal analysis circuitry 80 of processing circuitry 70 may determine, based on a time window of the plurality of digital samples, a gradient signal of a set of digital samples of the plurality of digital samples. Cardiac signal analysis circuitry 80 may determine, during the time window, a maximum point of the gradient signal.

[0067] In some examples, cardiac signal analysis circuitry 80 may determine a maximum point of the gradient signal by determining, based on detecting a peak of the filtered signal, the maximum point of the filtered signal. For examples, a second gradient (second derivative or slope of a slope) may be determined based on the filtered EGM signal. A zero crossing of a gradient signal of the filtered gradient signal may correspond in time with a maximum point of the filtered gradient signal. Cardiac signal analysis circuitry 80 may be configured to store on a memory device, for example memory 72, based on the maximum point, data indicating an occurrence of a T-wave end time on the filtered EGM signal. For example, cardiac signal analysis circuitry 80 may calculate the gradient signal by calculating a plurality of differences between adjacent samples of the plurality of digital samples. In some examples, cardiac signal analysis circuitry 80 may determine a maximum of the gradient signal, storing a time value corresponding to the maximum in memory 72 of IMD 16.

[0068] Cardiac signal analysis circuitry 80 may be further configured to measure one or more parameters of P-waves (e.g., atrial depolarization events). For example, cardiac signal analysis circuitry 80 may measure P-wave start time, P-wave end time, and P-wave duration, based on an area under a representative signal of the respective P-wave. Additionally, or alternatively, cardiac signal analysis circuitry 80 or processing circuitry 70 may measure P-R intervals (e.g., window of time separating P-wave start time and R-wave peak) and other intervals based on parameters measured using techniques of this disclosure. Cardiac signal analysis circuitry 80 may be configured calculate any combination of the derivative, integral, gradient, exponent, variance, amplitude, magnitude, and other mathematical operation of any portion of the cardiac EGM signal recorded by IMD 16.

[0069] Processing circuitry 70 and cardiac signal analysis circuitry 80 may be incorporated in a single processing unit or distributed among several processing units. Cardiac signal analysis circuitry 80 may be a component of, or a software or firmware module executed by, processing circuitry 70.

[0070] Telemetry circuitry 78 includes any suitable hardware, firmware, software or any combination thereof for communicating with another device, such as external device 24 (FIG. 1). Under the control of processing circuitry 70, telemetry circuitry 78 may receive downlink telemetry from and send uplink telemetry to external device 24 with the aid of an antenna, which may be internal and/or external. In some examples, processing circuitry 70 may transmit cardiac signals, e.g., EGM signals, produced by sensing circuitry 76.

[0071] Processing circuitry 70 may also generate and store marker codes indicative of different cardiac or other physiological events detected by sensing circuitry 76, processing circuitry 70, or cardiac signal analysis circuitry 80 and transmit the marker codes to external device 24. Information which processing circuitry 70 may transmit to external device 24 via telemetry circuitry 78 may also include an indication of a change in disease state of the heart, an indication of a change in heart response to the therapy provided or an indication that the heart continues to response in the same (or similar) manner to the therapy provided. Such information may be included as part of a marker channel with an EGM.

[0072] FIG. 4 is a block diagram illustrating example cardiac signal analysis circuitry 400 that may be used to identify one or more parameters of the cardiac EGM according to the techniques of this disclosure. Cardiac signal analysis circuitry 400 may be an example of cardiac signal analysis circuitry 80 of FIG. 3. Cardiac signal analysis circuitry 400 may include EGM pre-processing module 410, EGM signal 412, sampling rate 414, filtered EGM signal 416, QRS delineation module 420, QRS window start time 422, QRS window end time 424, QRS complex end time 426, T-wave delineation module 430, T-wave window start time 432, T-wave window end time 434, and T-wave end time 436. EGM pre-processing module 410, QRS delineation module 420, and T-wave delineation module 430 may be implemented within one or more microprocessors, DSPs, ASICs, FPGAs, or any other equivalent integrated or discrete logic QRS circuitry, as well as any combinations of such components. The term “module” may generally refer to any of the foregoing logic circuitry, alone or in combination with other logic circuitry, or any other equivalent circuitry, and alone or in combination with other digital or analog circuitry. Although the example illustrated in FIG. 4 describes analysis of a cardiac EGM signal sensed by implantable electrodes, the techniques described herein may be used for analysis of other cardiac signals, such as ECGs recorded by external electrodes.

[0073] EGM pre-processing module 710 may be configured to receive EGM signal 412 sensed by sensing circuitry 76 via electrodes of IMD 16. EGM signal 412 may represent a cardiac EGM signal including noise from various sources. Pre-processing module 410 may be configured to sample EGM signal 412 at sampling rate 414, producing a discrete EGM signal which may be modified using signal processing techniques described herein. In on example, EGM pre-processing module 410 may include a Butterworth filter defining a bandpass filter having a frequency range extending from 1 Hertz to 10 Hertz. Hence, low frequency noise and high frequency noise may be filtered out of the EGM signal by EGM pre-processing block 410. Examples of noise may include respiration activity, muscle activity, gastrointestinal activity, and noise from other systems in the body. Other examples of noise may include 60 Hertz noise, white noise, or other external noise at any frequency from external sources such as medical equipment or other machinery. Butterworth filters are infinite impulse response (IIR) filters that may be designed with a flat frequency response (e.g., the signal is attenuated equally among frequencies not passed by the filter). Additionally, or alternatively, EGM pre-processing module 410 may include any combination of HR filters, finite impulse response (FIR) filters, high-pass filters, low-pass filters, notch filters, Chebyshev filters, elliptic filters, Bessel filters, Gaussian filters, Linkwitz-Riley filters, or the like.

[0074] EGM pre-processing module 410 may produce filtered EGM signal 416. Since filters of EGM pre-processing module 410 are configured to remove noise from EGM signal 412, filtered EGM signal 416 may accentuate one or more features of a cardiac EGM signal previously obscured in EGM signal 412. For example, events such as P- waves, QRS complexes, and T-waves may be more visible in filtered EGM signal 316 than in EGM signal 412. Hence, parameters of a cardiac EGM may be more detectable in filtered EGM signal 416 than in EGM signal 412.

[0075] QRS delineation module 420 may receive filtered EGM signal 416 from EGM pre-processing module 410. Additionally, QRS delineation module 420 may sample filtered EGM signal 416 at sampling rate 414. QRS delineation module 420 may be configured to output QRS complex end time 426, the output representing a time in which a detected QRS complex concludes in the cardiac EGM signal. Initially, QRS delineation module 420 may set a QRS window indicative of a period of time in which a QRS complex is expected to end in a future portion of the cardiac EGM signal. For example, QRS delineation module 420 may detect an R-wave in filtered EGM signal 416 or receive an indication of an R-wave detection by other circuitry of IMD 16. Subsequently, QRS delineation module may set the QRS window beginning at QRS window start time 422 and extending to QRS window end time 424. In one example, QRS window start time 422 occurs 10 milliseconds following the detected R-wave, and QRS window end time 424 occurs 150 milliseconds following the detected R-wave. In other examples, QRS window start time 422 and QRS window end time 424 may include other values. QRS delineation module 420 may eliminate (e.g., “blank”) all portions of filtered EGM signal 416 outside of the QRS window.

[0076] After QRS delineation module 420 blanks filtered EGM 416 signal outside the QRS window, QRS delineation module 420 may calculate a gradient of filtered EGM signal 416 inside the QRS window (e.g., calculate the slope of filtered EGM signal 416 at each point throughout the QRS window). The gradient may be a real-time derivative calculation. In some examples, QRS delineation module 420 may define a representative signal as the gradient of filtered EGM signal 416. In other examples, QRS delineation module 420 may amplify the gradient of filtered EGM signal 416, creating the representative signal of the QRS complex. In other examples, QRS delineation module 420 may amplify the gradient of filtered EGM signal 416 by calculating an exponential (e.g., square, cube, or the like) of the gradient. The representative signal of the QRS complex may include a single polarity, meaning the polarity of the representative signal does not change throughout the QRS window. For example, the representative signal may include entirely negative or entirely positive values, however the representative signal does not alternate between positive and negative values.

[0077] QRS delineation module 420 may determine QRS complex end time 426, the output representing a time that a QRS complex in the cardiac EGM of the patient concludes. In calculating QRS complex end time 426, QRS delineation module 420 may evaluate an area under the representative signal of the QRS complex. For example, QRS delineation module 420 may determine the end of the QRS complex as an earliest determined time in which the area under the representative signal of the QRS complex is greater than or equal to a threshold portion of a total area under the representative signal of the QRS complex. The area under the representative signal of the QRS complex may be evaluated at each point along the representative signal. In one example, the threshold portion of the total area under the representative signal is 0.9.

[0078] Filtered EGM signal 416 may define a discrete signal having a plurality of data points spaced at sampling rate 414. Hence, the representative signal of the QRS complex may also define a discrete signal, and QRS delineation module 420 may compute the area under the representative signal at each data point of the plurality of data points. In one example, the area of any portion of the representative signal within the QRS window may be computed by performing a summation of data points beginning with a data point at the start of the QRS window and ending with a data point indicating an end of the respective portion of the representative signal of the QRS complex.

[0079] T-wave delineation module 430 may receive filtered EGM signal 416 from EGM pre-processing module 410. As T-wave delineation module 430 receives filtered EGM signal 416, the module may sample filtered EGM signal 416 at sampling rate 414. T-wave delineation module 430 may be configured to output T-wave end time 436. In particular, T- wave end time 436 represents a time in which a T-wave concludes in the cardiac EGM signal. Like QRS delineation module 420, T-wave delineation module 430 may set time windows. For example, T-wave delineation module 430 may set a T-wave window, the T- wave window indicative of a period of time in which a T-wave is expected to end in a future portion of the cardiac EGM signal. For example, T-wave delineation module 430 may detect an R-wave in filtered EGM signal 416. Subsequently, T-wave delineation module 430 may set the T-wave window beginning at T-wave window start time 422 and extending to T-wave window end time 424. In one example, T-wave window start time 432 occurs at QRS complex end time 426 calculated by QRS delineation module 420, and T-wave window end time 434 occurs 700 milliseconds following the detected R-wave. In other examples, T-wave window start time 432 and T-wave end time 434 may include other values. T-wave delineation module 430 may eliminate (e.g., “blank”) all portions of filtered EGM signal 416 outside of the T-wave window.

[0080] After T-wave delineation module 430 blanks filtered EGM signal 416 outside the T-wave window, T-wave delineation module 430 may calculate a gradient of filtered EGM signal 416 inside the T-wave window (e.g., calculate the slope of filtered EGM signal 416 at each point throughout the T-wave window). The gradient may be a real-time derivative calculation. In some examples, T-wave delineation module 430 may define a representative signal of the T-wave as the gradient of filtered EGM signal 416. In other examples, T-wave delineation module 430 may amplify the gradient of filtered EGM signal 416, creating the representative signal of the T-wave. In other examples, T-wave delineation module 430 may amplify the gradient of filtered EGM signal 416 by calculating an exponential (e.g., square, cube, or the like) of the gradient. The representative signal of the T-wave may include a single polarity, meaning the polarity of the representative signal does not change throughout the T-wave window. For example, the representative signal may include entirely negative or entirely positive values, however the representative signal does not alternate between positive and negative values.

[0081] T-wave delineation module 430 may determine T-wave end time 436. In calculating T-wave end time 436, T-wave delineation module 430 may be configured to evaluate an area under the representative signal of the T-wave. For example, T-wave delineation module 430 may determine the end of the T-wave as an earliest determined time in which the area under the representative signal of the T-wave is greater than or equal to a threshold portion of a total area under the representative signal of the T-wave. The area under the representative signal of the T-wave may be evaluated at each point along the representative signal. In one example, the threshold portion of the total area under the representative signal is 0.6. The representative signal of the T-wave may be a discrete signal, and an area under any portion of the representative signal may be determined by performing a summation of all data points within the respective portion of the representative signal. T- wave delineation module 430 may be configured to blank a representative signal. T-wave delineation module 430 may calculate a gradient signal from the blanked signal a determine a maximum point using a peak detection technique.

[0082] FIG. 5 is a voltage/time graph illustrating a plot of an example 500 of an EGM signal 514 plotted over time, in accordance with one or more techniques of this disclosure. EGM signal 514 is plotted as a line plot having an abscissa axis 512, representing time, and a first ordinate axis 510, representing an electrical voltage amplitude. EGM signal 514 may be an example of EGM signal 412 of FIG. 4. EGM signal 514 may include portions where the voltage amplitude is positive (e.g., above the abscissa axis), zero, or negative (e.g., below the abscissa axis). The electrical voltage amplitude represented by EGM signal 514 may be a voltage measured by a sensor, for example sensor(s) 82 of FIG. 3.

[0083] In some examples, EGM signal 514 may represent an EGM signal processed by a processing circuitry of an implantable medical device. EGM signal 514 may be an example of EGM signal 412 of FIG. 4. EGM signal 514 may include signal features having a variety of slopes and durations. Some signal features may be filtered out and some features may be further processed for detection or analysis.

[0084] In some examples, signal features occurring during a portion 520 of EGM signal 514 may cause errors when determining the occurrence of a T-wave. For example, signal features have large positive or negative amplitudes may increase the likelihood of false detection or distortion in processing the EGM signal. In some examples, EGM signal 514 may be processed with a blanking technique to remove signal features that may cause false detections or errors. Processing circuitry, for example cardiac signal analysis circuitry 400, may further process a signal using a gradient technique.

[0085] FIG. 6 is a voltage/time graph illustrating an example 600 of an EMG signal 616 before and after applying a blanking and gradient techniques, according to the techniques of this disclosure. Filtered EGM signal 616 represents an EGM signal before applying a blanking processing technique or a gradient processing technique. Gradient signal 618 represents an EGM signal after applying both a blanking processing technique and a gradient processing technique. Both filtered EGM signal 616 and gradient signal 618 are plotted as line plots having an abscissa axis 612, representing time, and an ordinate axis 610, representing an electrical voltage amplitude.

[0086] Filtered EGM signal may be a representative signal derived from a cardiac signal sensed by sensing circuitry 76. Cardiac processing circuitry 400 for undesired frequency noise including sixty-hertz noise, respiratory system generated noise, muscular system generated noise, digestive system generated noise, nervous system generated noise, and other noise generated by biological systems within patient 14, and external noise generated by external sources such as medical equipment or other machinery.

[0087] Processing fdtered EGM signal 616, with a blanking technique and a gradient technique, into a gradient signal 618, may be performed by cardiac processing circuitry 400. Cardiac processing circuitry 400 may apply the blanking technique before, or after applying the gradient technique. In some examples, cardiac processing circuitry 400 may apply the blanking technique concurrently with the gradient technique.

[0088] In some examples, T-wave delineation module 430 of cardiac processing circuitry may use a blanking technique while processing fdtered EGM signal 616 into gradient EGM signal 618. For example, the time window may correspond to the set of digital samples having a predetermined relationship to a sample of the plurality of digital samples. The set of digital samples, of a plurality of digital samples, may represent samples of fdtered EGM signal 616 outside the time window. T-wave delineation module 430 may be configured to blank samples outside the time window. After T-wave delineation module 430 performs a blanking technique, gradient signal 618 may level (e.g., time fdter, blank, clear, erase) a first portion 620A and a second portion 620B. For example, blanking first portion 620A and second portion 620B may be performed by setting an amplitude of first portion 620A and second portion 620B of EGM signal 614 to a constant value (e.g., zero). In some examples, T-wave delineation module 430 may blank filtered EGM signal 616 by limiting a time duration, for example T-wave window, over which further processing techniques are applied. [0089] In some examples, the time window may correspond to the set of digital samples having a predetermined relationship to a sample of the plurality of digital samples. In various examples, the time window may include adjacent samples determined during a fixed time duration, and wherein a center of the time window corresponds to a middle sample of the set of digital samples. Alternatively, the time window may be determined based on a dynamic time duration. For examples, the time window may correspond to samples determined during a time duration, and T-wave delineation module 430 may determine the time duration based on a heart rate represented by the cardiac signal.

[0090] T-wave delineation module 430 may apply the gradient technique while processing filtered EGM signal 616 into gradient EGM signal 618. Applying the gradient technique to filtered EGM signal 616, a first representative signal, resulting in gradient signal 618 a second representative signal, T-wave delineation module 430 may use digital signal processing techniques to calculate a gradient (e.g., first derivative, slope, or difference between samples). For example, digital samples over first portion 620A, second portion 620B, and windowed portion 622. T-wave delineation module 430 may apply the gradient technique by calculating a plurality of differences between adjacent samples of the plurality of digital samples. When applying the gradient technique after the blanking technique, the differences between adjacent sample may be calculated exclusively within T-wave window 622. For examples, when determining filtered EGM signal 616, T-wave delineation module 430 may determine a plurality of differences between adjacent digital samples within the time window. In examples where the gradient technique is applied before the blanking technique, differences between adjacent samples may be calculated over first portion 620A, second portion 620B, and T-wave window.

[0091] In various examples, T-wave delineation module 430 may filter gradient signal 618 with a frequency filter, to remove undesired frequency content. For example, T-wave delineation module 430 may filter gradient signal 618 with a bandpass filter. In some examples the bandpass filter by have a passband from 1Hz to 20Hz. Filtering gradient signal 618 may be the same as, or similar to, filtering the slope of filtered EGM signal 616. Filtering gradient signal 618 may reduce errors in determining a peak value of the gradient signal 618.

[0092] While determining T-wave end time 436, T-wave delineation module 430 may calculate a maximum point 630A of filtered gradient signal 618. In various examples, maximum point 630 A may be determined after gradient signal 618 has been filtered with the bandpass filter. For example, T-wave delineation module 430 may apply a peak detection technique to gradient signal 618 after being filtered with the bandpass filter. The peak detection technique may include calculating a zero crossing of a second derivative. For example, T-wave delineation module 430 may calculate a gradient (a second derivative, a slope, or a difference between adjacent samples) of gradient signal 618 resulting in a second derivative signal (not illustrated). Determining a time value 630B, corresponding to a zero amplitude of the second derivative signal, may result in determining time value 630B at which gradient signal 618 has a maximum amplitude.

[0093] FIG. 7 is a voltage/time graph illustrating an example 700 of a squared gradient signal 720 calculated based on a gradient signal 718, according to the techniques of this disclosure. Gradient signal 718 may be an example of gradient signal 618 of Fig. 6. Gradient signal 718 may be a representative signal of a cardiac signal after a gradient technique and a blanking technique have been applied. Squared gradient signal 720 may represent an exponential (e.g., square, cube, or the like) of an EGM signal 716. For example, as illustrated, squared gradient signal 720 may be a square of gradient signal 718. Both gradient signal 718 and squared gradient signal 720 are plotted as line plots having an abscissa axis 712, representing time, and an ordinate axis 710, representing an electrical voltage amplitude. [0094] In some examples, T-wave delineation module 430 may determine, based on gradient signal 718, squared gradient signal 720. For example, each sample of a plurality of samples representing gradient signal 718, may be squared (e.g., raised to the second exponential) resulting in a second plurality of samples representing squared gradient signal 720. In some examples, the exponential used in calculating squared gradient signal 720 may be an even exponential, resulting in a single polarity signal.

[0095] T-wave delineation module 430, or other processing circuitry of the IMD, may determine, based on squared gradient signal 720, a programmable energy point 730A by an area under squared gradient signal 720. For examples, programmable energy point 730A may correspond to a time at which an area under gradient squared signal 720 from a beginning of gradient signal 720 to the time is a portion of the total area under squared signal 720. The portion may be a percentage, for example 65%, 60%, or 55%, of the total area. In examples when a blanking technique is applied, T-wave delineation module 430 may calculate a total area under squared signal 720 within T-wave window 722. T-wave delineation module 430 A, may determine a time value 730B corresponding to programmable energy point 730 A by calculating a first area under squared signal 720 from a start of T-wave window 722 to a first time value. T-wave delineation module 430A may divide the first area under squared signal 720 by the total area under squared signal 720 resulting in a percentage. If the percentage is equal to a predetermine energy percentage, then the first time value is time value 730.

[0096] In some examples, the time window may include adjacent samples determined during a dynamic time duration based on the programmable energy point. T-wave delineation module 430, may determine the time window based on the programmable energy point. For example, a fixed time duration may be used to define a length of the time window centered around the programmable energy point. Centering the time window around the programmable energy point may limit the length of the time window used to determine a maximum value.

[0097] FIG. 8 is a zoomed in voltage/time graph illustrating an example 800 comparison between a first time value 820 determined based on a programmable energy point and a second time value 822 determined based on a maximum point, according to the techniques of this disclosure. EGM signal 816 and gradient signal 818 may be a zoomed in representations of EGM signal 616 gradient signal 618 of FIG 6. Both EGM signal 816 and gradient signal 818 are plotted as line plots having an abscissa axis 812, representing time, and an ordinate axis 810, representing an electrical voltage amplitude.

[0098] In some examples, first time value 820 may include a time value determined by cardiac processing circuitry 400 corresponding to a maximum point of gradient signal 818. Cardiac processing circuitry 400 may determine first time value 820 based on determining the maximum point using any of the maximum point techniques discussed in FIGs 1-7. [0099] In some examples, second time value 822 may include a time value determined by cardiac processing circuitry 400 corresponding to a programmable energy point of squared gradient signal 720. Cardiac processing circuitry 400 may determine second time value 822 based on determining the programable energy point using one or more techniques discussed in FIGs 1-7.

[0100] FIG. 9 is a flow diagram illustrating an example 900 technique for determining a time value corresponding to an occurrence of a T-wave end time, according to the techniques of this disclosure. Processing circuitry of the IMD may be configured to determine, by the processing circuitry based on the cardiac signal sensed by the sensing circuitry of the medical device, a digital signal comprising a plurality of digital samples representing the cardiac signal (910). Processing circuitry may include cardiac processing circuitry 400 and more specifically T-wave delineation module 430. Cardiac processing circuitry 400 may use sensing circuitry 76 to sense a cardiac signal and process the cardiac signal with an ADC into a plurality of digital samples.

[0101] The processing circuitry may be configured to determine, by the processing circuitry and based on a time window of the plurality of digital samples, a gradient signal of a set of digital samples (920). The plurality of digital samples may include digital samples produced by the ADC in the presence of a cardiac signal. To determine the gradient signal, the processing circuitry may use any of the techniques for calculating a gradient signal discussed in FIGs 1-9.

[0102] The processing circuitry may be configured to determine, by the processing circuitry and during the time window, a maximum point of the gradient signal (930). The maximum point may include maximum point 630A of FIG 6. The programmable energy point may include programmable energy point 730A of FIG. 7.

[0103] The processing circuity may be configured to store on a memory device, by the processing circuitry and based on the maximum point, data indicating an occurrence of a T- wave end time (940). The memory device may include memory 72. More specifically, memory 72 may store data indicating an occurrence of a T-wave end time in T-wave window parameter 88 of memory 72.

[0104] FIG. 10 is a flow diagram illustrating an example 1000 technique for determining a time value using a maximum technique based on a filtered ECG signal, according to the techniques of this disclosure. Example 100 may include a technique for determining a maximum point that may be implemented by IMD 16. Processing circuitry of the IMD may be configured to blank, outside of a T-wave window, a filtered ECG signal resulting in a blanked signal (1010). Processing circuitry may include cardiac processing circuitry 400 and more specifically T-wave delineation module 430.

[0105] Processing circuitry may be configured to calculate, based on the blanked signal a gradient signal (1020). The gradient signal may include any of gradient signal(s) 618, 718, or 818 of FIGs 6-8. The blanked signal may include ECG signal 616 blanked over t-window 622 of FIG. 6.

[0106] Processing circuitry may be configured to filter, with a bandpass filter, the gradient signal (1030). The gradient signal may include any of gradient signal(s) 618, 718, or 818 of FIGs 6-8. Processing circuitry may filter, using a 1Hz to 20Hz bandpass filter, any of gradient signal(s) 618, 718, or 818 of FIGs 6-8. Processing circuitry may use any of filter configures discussed in FIG. 4 to filter the gradient signal.

[0107] Processing circuitry may be configured to determine a time value corresponding to a maximum amplitude of the filtered gradient signal (1040). The maximum amplitude may include maximum point 630A of FIG. 6 of maximum point 820 of FIG. 8. The time value may include first time value 820 of FIG. 8.

[0108] The techniques described in this disclosure may be implemented, at least in part, in hardware, software, firmware, or any combination thereof. For example, various aspects of the techniques may be implemented within one or more microprocessors, DSPs, ASICs, FPGAs, or any other equivalent integrated or discrete logic QRS circuitry, as well as any combinations of such components, embodied in external devices, such as physician or patient programmers, stimulators, or other devices. The terms “processor,” “processing circuitry,” “controller” or “control module” may generally refer to any of the foregoing logic circuitry, alone or in combination with other logic circuitry, or any other equivalent circuitry, and alone or in combination with other digital or analog circuitry.

[0109] For aspects implemented in software, at least some of the functionality ascribed to the systems and devices described in this disclosure may be embodied as instructions on a non-transitory computer-readable storage medium such as RAM, ROM, NVRAM, EEPROM, FLASH memory, magnetic media, optical media, or the like. The instructions may be executed to support one or more aspects of the functionality described in this disclosure. [0110] Various examples have been described. These and other examples are within the scope of the following claims.

[OHl] The following examples may illustrate one or more aspects of the disclosure.

[0112] Example 1. A method comprising: determining, by processing circuitry of a medical device system and based on a cardiac signal sensed by a medical device of the medical device system, a digital signal comprising a plurality of digital samples representing the cardiac signal; determining, by the processing circuitry and based on a time window of the plurality of digital samples, a gradient signal of a set of digital samples of the plurality; determining, by the processing circuitry and during the time window, a maximum point of the gradient signal; and storing on a memory device, by the processing circuitry and based on the maximum point, data indicating a T-wave end time.

[0113] Example 2. The method of example 1, wherein the set of digital samples represents a time window and wherein determining the gradient signal comprises: blanking samples outside the time window; and determining a plurality of differences between adjacent digital samples within the time window.

[0114] Example 3. The method of examples 1 or 2, wherein determining the maximum point of the gradient signal comprises: fdtering, by a bandpass fdter, the gradient signal into a fdtered signal; determining, based on detecting a peak of the filtered signal, the maximum point of the filtered signal.

[0115] Example 4. The method of examples 1-3, wherein determining the maximum of the gradient signal comprises: determining, based on the gradient signal, a squared gradient signal; and determining, based on the squared gradient signal, a programmable energy point by an area under the squared gradient signal.

[0116] Example 5. The method of examples 1-4, wherein the time window corresponds to the set of digital samples having a predetermined relationship to a sample of the plurality of digital samples.

[0117] Example 6. The method of examples 1-5, wherein the time window comprises adjacent samples determined during a fixed time duration, and wherein a center of the time window corresponds to a middle sample of the set of digital samples. [0118] Example 7. The method of examples 1-6, wherein the time window corresponds to samples determined during a time duration, and the method comprises determining the time duration based on a heart rate represented by the cardiac signal.

[0119] Example 8. The method of examples 1-7, wherein determining the time window comprises: determining, based on the gradient signal, a squared gradient signal; determining, based on the squared gradient signal, the programmable energy point by an area under the squared gradient signal; and determining the time window based on the programmable energy point.

[0120] Example 9. A medical device comprising: a memory device; sensing circuitry configured to sense a cardiac signal via a plurality of electrodes; and processing circuitry configured to: determine, by the processing circuitry based on the cardiac signal sensed by the sensing circuitry of the medical device, a digital signal comprising a plurality of digital samples representing the cardiac signal; determine, by the processing circuitry and based on a time window of the plurality of digital samples, a gradient signal of a set of digital samples; determine, by the processing circuitry and during the time window, a maximum point of the gradient signal; and store on the memory device, by the processing circuitry and based on the maximum point, data indicating a T-wave end time.

[0121] Example 10. The medical device of example 9, wherein to determine the gradient signal the processing circuitry is further configured to: blank samples outside the time window; and determine a plurality of differences between amplitudes of the plurality of digital samples within the time window.

[0122] Example 11. The medical device of examples 9 or 10, wherein to determine the maximum point of the gradient signal, the processing circuitry is further configured to: filter, by a bandpass filter, the gradient signal into a filtered signal; determine, based on a peak of the filtered signal, the maximum point of the filtered signal.

[0123] Example 12. The medical device of examples 9-11, wherein to determine the time value, the processing circuitry is further configured to: determine, based on the gradient signal, a squared gradient signal; and determine, based on the squared gradient signal, a programmable energy point of the squared gradient signal.

[0124] Example 13. The medical device of examples 9-12, wherein the time window corresponds to the set of digital samples and wherein the set of digital samples has a predetermined relationship to a sample of the plurality of digital samples.

[0125] Example 14. The medical device of examples 9-13, wherein the time window comprises adjacent samples determined during a fixed time duration, and wherein a center of the time window corresponds to a middle sample of the set of digital samples.

[0126] Example 15. The medical device of examples 9-14, wherein the time window corresponds to samples determined during a time duration, and the time duration based on a heart rate represented by the cardiac signal.

[0127] Example 16. The medical device of examples 9-15, wherein to determine the time window the processing circuitry is configured to: determine, based on the gradient signal, a squared gradient signal; determine, based on the squared gradient signal, the programmable energy point by an area under the squared gradient signal; and determine the time window based on the programmable energy point.

[0128] Example 17. The medical device of examples 9-16, wherein the medical device comprises an implantable medical device.

[0129] Example 18. The medical device of examples 9-17, wherein the medical device comprises at least one of a pacemaker, a cardioverter, or a defibrillator.

[0130] Example 19. A system comprising: an implantable medical device configured to sense a cardiac signal; and processing circuitry configured to: determine, by the processing circuitry and based on the cardiac signal sensed by the implantable medical device of the system, a digital signal comprising a plurality of digital samples representing the cardiac signal; determine, by the processing circuitry and based on a time window of the plurality of digital samples, a gradient signal of a set of digital samples; determine, by the processing circuitry and during the time window, a maximum point of the gradient signal; and store on a memory device of the implantable medical device, by the processing circuitry and based on the maximum point, data indicating a T-wave end time.

[0131] Example 20. The system of example 19, wherein to determine the gradient signal the processing circuitry is further configured to: blank samples outside the time window; and determine a plurality of differences between amplitudes of the plurality of digital samples with the time window.

[0132] Example 21. The system of examples 19 or 20, wherein to determine the maximum point of the gradient signal, the processing circuitry is further configured to: filter, by a bandpass filter, the gradient signal into a filtered signal; determine, based on a peak of the filtered signal, the maximum point of the filtered signal.

[0133] Example 22. The system of examples 19-21, wherein to determine the time value, the processing circuitry is further configured to: determine, based on the gradient signal, a squared gradient signal; and determine, based on the squared gradient signal, a programmable energy point of the squared gradient signal.

[0134] Example 23. The system of claim 19-22, wherein the time window corresponds to the set of digital samples and wherein the set of digital samples has a predetermined relationship to a sample of the plurality of digital samples.

[0135] Example 24. The system of claim 19-23, wherein the time window comprises adjacent samples determined during a time duration, and the method comprises determining the time duration based on a heart rate represented by the cardiac signal.

Claims

What is claimed is:

1. A medical device comprising: a memory device; sensing circuitry configured to sense a cardiac signal via a plurality of electrodes; and processing circuitry configured to: determine, by the processing circuitry based on the cardiac signal sensed by the sensing circuitry of the medical device, a digital signal comprising a plurality of digital samples representing the cardiac signal; determine, by the processing circuitry and based a time window of the plurality of digital samples, a gradient signal of a set of digital samples; determine, by the processing circuitry and during the time window, a maximum point of the gradient signal; and store on the memory device, by the processing circuitry and based on the maximum point, data indicating a T-wave end time.

2. The medical device of claim 1, wherein to determine the gradient signal the processing circuitry is further configured to: blank samples outside the time window; and determine a plurality of differences between amplitudes of the plurality of digital samples within the time window.

3. The medical device of claim 1 or 2, wherein to determine the maximum point of the gradient signal, the processing circuitry is further configured to: filter, by a bandpass filter, the gradient signal into a filtered signal; determine, based on a peak of the filtered signal, the maximum point of the filtered signal.

4. The medical device of any one or more of claims 1 to 3, wherein to determine the time value, the processing circuitry is further configured to: determine, based on the gradient signal, a squared gradient signal; and determine, based on the squared gradient signal, a programmable energy point of the squared gradient signal.

5. The medical device of any one or more of claims 1 to 4, wherein the time window corresponds to the set of digital samples and wherein the set of digital samples has a predetermined relationship to a sample of the plurality of digital samples.

6. The medical device of any one or more of claims 1 to 5, wherein the time window comprises adjacent samples determined during a fixed time duration, and wherein a center of the time window corresponds to a middle sample of the set of digital samples.

7. The medical device of any one or more of claims 1 to 6, wherein the time window corresponds to samples determined during a time duration, and the time duration is based on a heart rate represented by the cardiac signal.

8. The medical device of any one or more of claims 1 to 7, wherein to determine the time window the processing circuitry is configured to: determine, based on the gradient signal, a squared gradient signal; determine, based on the squared gradient signal, the programmable energy point by an area under the squared gradient signal; and determine the time window based on the programmable energy point.

9. The medical device of any one or more of claims 1 to 8, wherein the medical device comprises an implantable medical device.

10. The medical device of any one or more of claims 1 to 9, wherein the medical device comprises at least one of a pacemaker, a cardioverter, or a defibrillator.

11. A system comprising: an implantable medical device configured to sense a cardiac signal; and processing circuitry configured to: determine, by the processing circuitry and based on the cardiac signal sensed by the implantable medical device of the system, a digital signal comprising a plurality of digital samples representing the cardiac signal; determine, by the processing circuitry and based on a time window of the plurality of digital samples, a gradient signal of a set of digital samples; determine, by the processing circuitry and during the time window, a maximum point of the gradient signal; and store on a memory device of the implantable medical device, by the processing circuitry and based on the maximum point, data indicating a T-wave end time.

12. The system of claim 11, wherein to determine the gradient signal the processing circuitry is further configured to: blank samples outside the time window; and determine a plurality of differences between amplitudes of the plurality of digital samples with the time window.

13. The system of claim 11 or 12, wherein to determine the maximum point of the gradient signal, the processing circuitry is further configured to: filter, by a bandpass filter, the gradient signal into a filtered signal; determine, based on a peak of the filtered signal, the maximum point of the filtered signal.

14. The system of any one or more of claims 11 to 13, wherein to determine the time value, the processing circuitry is further configured to: determine, based on the gradient signal, a squared gradient signal; and determine, based on the squared gradient signal, a programmable energy point of the squared gradient signal.

15. The system of any one or more of claims 11 to 14, wherein the time window corresponds to the set of digital samples and wherein the set of digital samples has a predetermined relationship to a sample of the plurality of digital samples.