PVC onset detection and storage

Abstract

A method includes detecting a first premature ventricular contraction (PVC) episode occurring within a first length of electrocardiogram (ECG) data. The method also includes recording the first length of ECG data into a memory in response to the detecting. A second PVC episode is determined to have occurred after the first PVC episode and within a second length of ECG data, and the second PVC episode is determined to be a first type of PVC episode or a second type of PVC episode. The method also includes determining whether to overwrite the first length of ECG data or to separately record the second length of ECG data into the memory based at least in part on the determining whether the second PVC episode is the first type of PVC episode or the second type of PVC episode.

Description

[0001] Cross-references to related applications

[0002] This application claims priority to provisional application No. 63 / 551,516, filed on February 8, 2024, the entire contents of which are incorporated herein by reference. Technical Field

[0003] Examples of this disclosure generally relate to methods for the efficient recording of cardiac activity data involving premature ventricular contractions. Background Technology

[0004] Medical devices that allow doctors to monitor heart activity are becoming increasingly common in diagnosing and treating patients' medical conditions. For example, cardiac monitoring can be used to identify abnormal heart rhythms, providing critical alerts to patients, doctors, or other caregivers, and enabling treatment as needed. Summary of the Invention

[0005] In Example 1, a method includes detecting a first premature ventricular contraction (PVC) within a first length of electrocardiogram (ECG) data. The first PVC is a first type of PVC. The method further includes, in response to the detection, recording the first length of ECG data into a memory. A second PVC is determined to have occurred after the first PVC and within a second length of ECG data, and the second PVC is determined to be either a first type of PVC or a second type of PVC. The method further includes, at least in part, determining whether to overwrite the first length of ECG data or to record the second length of ECG data separately into the memory based on whether the second PVC is a first type of PVC or a second type of PVC.

[0006] In Example 2, the method according to Example 1 further includes determining that the second PVC attack is the first type of PVC attack, and in response, overwriting the first length of ECG data with ECG data of the second length.

[0007] In Example 3, according to the method of any one of Example 1 or 2, wherein the first type of PVC outbreak and the second type of PVC outbreak are maximum load outbreaks, wherein the first PVC outbreak is associated with a first load value, wherein the second PVC outbreak is associated with a second load value, wherein in response to the second load value being greater than the first load value, the first length of ECG data is covered by the second length of ECG data.

[0008] In Example 4, the method according to Example 3 further includes detecting that a third PVC occurs after the second PVC and within a third length of ECG data. The third PVC is associated with a third load value. The method further includes, after determining that the third load value is greater than the second load value, overwriting the second length of ECG data with the third length of ECG data.

[0009] In Example 5, the method described in Example 3 or 4 is used, wherein the first load value and the second load value are based on the number of PVCs within a predetermined time period.

[0010] In Example 6, according to the method of any one of Examples 1 or 2, wherein the first type of PVC outbreak and the second type of PVC outbreak are paired events and / or triple events, wherein the first PVC outbreak is associated with a first number of pairs or triples, wherein the second PVC outbreak is associated with a second number of pairs or triples, wherein, in response to the second number of pairs or triples being greater than the first number of pairs or triples, the first length of ECG data is covered by the second length of ECG data.

[0011] In Example 7, according to the method of Example 1, the first type and the second type are different types of PVC outbreaks. The method further includes: determining that the second PVC outbreak is a PVC outbreak of the second type; and in response, recording ECG data of the second length into the memory, such that the memory contains both the ECG data of the first length and the ECG data of the second length.

[0012] In Example 8, according to the method of Example 7, wherein the first type of PVC seizure is a maximum load seizure, and wherein the second type of PVC seizure is a paired seizure or a triple seizure.

[0013] In Example 9, according to the method of any one of Examples 1-8, the detection of the occurrence of the first PVC outbreak is based at least in part on the satisfaction of a minimum PVC threshold.

[0014] In Example 10, the method according to any one of Examples 1-9 further includes temporarily storing the first length in a buffer before recording the ECG data of the first length into the memory.

[0015] In Example 11, according to the method described in Example 10, the ECG data not recorded in the memory is permanently deleted.

[0016] In Example 12, the method according to any one of Examples 1-11 further includes wirelessly transmitting any PVC outbreaks recorded in the memory to a remote computing device.

[0017] In Example 13, a computer program product includes instructions that cause one or more processors to perform the steps of the method described in Examples 1-12.

[0018] In Example 14, a computer-readable medium is provided on which a computer program product according to Example 13 is stored.

[0019] In Example 15, an implantable medical device includes a computer-readable medium as described in Example 14.

[0020] In Example 16, the medical device includes a housing and electrodes coupled to the housing. The electrodes are used to sense cardiac activation signals for generating ECG data. The medical device also includes a memory and one or more processors programmed with instructions to cause the medical device to: (1) detect a first PVC attack occurring within a first length of ECG data, wherein the first PVC attack is a first type of PVC attack; (2) in response to the detection, record the first length of ECG data into the memory; (3) detect a second PVC attack occurring after the first PVC attack and within a second length of ECG data; (4) determine whether the second PVC attack is a first type of PVC attack or a second type of PVC attack; and (5) determine, at least in part, based on whether the second PVC attack is a first type of PVC attack or a second type of PVC attack, whether to overwrite the first length of ECG data or to record the second length of ECG data separately into the memory.

[0021] In Example 17, the medical device according to Example 16 further includes a buffer, wherein the ECG data of the first length is temporarily stored in the buffer before being recorded into the memory.

[0022] In Example 18, according to the medical device of Example 17, the ECG data temporarily stored in the buffer but not recorded in the memory is permanently deleted.

[0023] In Example 19, the medical device according to Example 16, wherein the one or more processors are programmed with instructions to further cause the medical device to determine that the second PVC attack is the first type of PVC attack, and in response, to overwrite the first length of ECG data with ECG data of the second length.

[0024] In Example 20, the medical device according to Example 16, wherein the first type of PVC attack and the second type of PVC attack are maximum load attacks, wherein the first PVC attack is associated with a first load value, wherein the second PVC attack is associated with a second load value, wherein in response to the second load value being greater than the first load value, the first length of ECG data is covered by the second length of ECG data.

[0025] In Example 21, the medical device according to Example 20 is used, wherein the first load value and the second load value are based on the number of PVCs within a predetermined time period.

[0026] In Example 22, according to the medical device of Example 16, wherein the first type of PVC attack and the second type of PVC attack are paired events and / or triple events, wherein the first PVC attack is associated with a first number of pairs or triples, wherein the second PVC attack is associated with a second number of pairs or triples, wherein, in response to the second number of pairs or triples being greater than the first number of pairs or triples, the first length of ECG data is covered by the second length of ECG data.

[0027] In Example 23, according to the medical device of Example 16, the first PVC outbreak is detected at least in part based on meeting a minimum PVC threshold.

[0028] In Example 24, the medical device according to Example 16, wherein the one or more processors are programmed with instructions to further cause the medical device to wirelessly transmit any PVC episodes recorded in the memory to a remote computing device at periodic intervals.

[0029] In Example 25, the medical device according to Example 16, wherein the memory includes a long-term memory and a temporary memory, wherein ECG data of the first length is first recorded to the temporary memory and then recorded to the long-term memory.

[0030] In Example 26, the medical device according to Example 25 is provided, wherein the storage capacity of the long-term memory is greater than the storage capacity of the temporary memory.

[0031] In Example 27, the medical device according to Example 25 is wherein the long-term memory is flash memory and the temporary memory is random access memory.

[0032] In Example 28, the medical device according to Example 25 is wherein the temporary memory is a first-in-first-out (FIFO) memory.

[0033] In Example 29, a method includes detecting a first PVC outbreak occurring within a first length of ECG data. The first PVC outbreak is a first type of PVC outbreak. In response to the detection, the first length of ECG data is recorded into memory. The method further includes detecting a second PVC outbreak occurring after the first PVC outbreak and within a second length of ECG data; determining whether the second PVC outbreak is a first type of PVC outbreak or a second type of PVC outbreak; and determining, at least in part, based on the determination that the second PVC outbreak is a first type of PVC outbreak or a second type of PVC outbreak, whether to overwrite the first length of ECG data or to record the second length of ECG data separately into the memory.

[0034] In Example 30, the method according to Example 29 further includes determining that the second PVC attack is the first type of PVC attack; and in response, overwriting the first length of ECG data with ECG data of the second length.

[0035] In Example 31, according to the method of Example 29, wherein the first type of PVC outbreak and the second type of PVC outbreak are maximum load outbreaks, wherein the first PVC outbreak is associated with a first load value, wherein the second PVC outbreak is associated with a second load value, wherein in response to the second load value being greater than the first load value, the first length of ECG data is covered by the second length of ECG data.

[0036] In Example 32, according to the method of Example 31, the first load value and the second load value are based on the number of PVCs within a predetermined time period.

[0037] In Example 33, according to the method of Example 29, wherein the first type of PVC outbreak and the second type of PVC outbreak are paired events and / or triple events, wherein the first PVC outbreak is associated with a first number of pairs or triples, wherein the second PVC outbreak is associated with a second number of pairs or triples, wherein, in response to the second number of pairs or triples being greater than the first number of pairs or triples, the first length of ECG data is covered by the second length of ECG data.

[0038] In Example 34, according to the method of Example 29, the first type and the second type are different types of PVC attacks. The method further includes: determining that the second PVC attack is a PVC attack of the second type; and in response, recording ECG data of the second length into the memory, such that the memory contains both the ECG data of the first length and the ECG data of the second length.

[0039] In Example 35, according to the method of Example 34, wherein the first type of PVC seizure is a maximum load seizure, and wherein the second type of PVC seizure is a paired seizure or a triple seizure.

[0040] While several examples have been disclosed, other examples of this disclosure will become apparent to those skilled in the art from the following detailed description, which illustrates and describes illustrative examples of this disclosure. Therefore, the accompanying drawings and detailed description should be considered illustrative in nature, rather than restrictive. Attached Figure Description

[0041] Figure 1 This is a schematic diagram of a system including a medical device according to certain examples of this disclosure.

[0042] Figure 2-4 Different views of medical devices according to certain examples of this disclosure are shown.

[0043] Figure 5 A block diagram depicting an illustrative method according to certain examples of this disclosure is shown.

[0044] Figure 6 ECG data bars according to certain examples of this disclosure are shown.

[0045] Figure 7 This is a block diagram illustrating an illustrative computing device based on examples of this disclosure.

[0046] While the disclosed subject matter can be readily modified and alternatively formulated in various ways, specific examples are shown by way of example in the accompanying drawings and described in detail below. However, it is not intended to limit the disclosed subject matter to the specific examples described. Rather, the disclosed subject matter is intended to cover all modifications, equivalents, and alternatives that fall within the scope of the disclosed subject matter as defined by the appended claims. Detailed Implementation

[0047] Medical devices can be equipped with one or more sensing components (e.g., sensors, electrodes) and programmed to sense physiological data, such as electrocardiogram (ECG) data. To collect physiological data, one or more medical devices (e.g., implantable cardiac monitors / recorders, external cardiac monitors / recorders) can be implanted in or coupled to a patient, enabling the device to sense physiological data. While such a medical device may attempt to continuously sense physiological data (such as ECG data), to conserve power and storage capacity, it may ultimately not record all sensed physiological data and transmit all sensed physiological data externally. Instead, sensed physiological data may only be recorded and then transmitted under certain circumstances. For example, a medical device can be programmed to process and analyze sensed physiological data to determine the occurrence of certain health events, such as the occurrence of premature ventricular contractions (PVCs). If a PVC is determined to have occurred, some sensed physiological data can be recorded in the medical device's long-term memory and then transmitted to another device. However, recording and transmitting physiological data every time a PVC occurs can be an inefficient use of the medical device's power and storage capacity.

[0048] Therefore, some instances of this disclosure relate to methods for prioritizing which cardiac events involving PVCs are recorded and ultimately transmitted for viewing.

[0049] Cardiac Event Assessment System

[0050] Figure 1 This is a schematic diagram of a medical device 100 within a cardiac event assessment system. The medical device 100 can be subcutaneously implanted in an implantation site or pocket in the patient's chest or abdomen and can be configured to sense physiological signals associated with the patient's heart 12. The medical device 100 can be an implantable cardiac monitor (e.g., an implantable diagnostic monitor, an implantable loop recorder) configured and programmed to record physiological parameters such as one or more cardiac activation signals, heart sounds, blood pressure measurements, and blood oxygen saturation. Although... Figure 1 The medical device 100 is shown as an implantable cardiac monitor, but the methods described herein can be used in conjunction with other types of implantable medical devices, such as pulse generators (e.g., pacemakers, defibrillators). Furthermore, although Figure 1 The medical device 100 is shown as a medical device implanted in the body of the patient 10, but the methods described herein can be used in conjunction with external medical devices, such as monitoring devices coupled to the patient’s skin or mobile devices including one or more sensors for detecting physiological data and / or capable of processing the detected physiological data.

[0051] exist Figure 1 In the example, medical device 100 includes electrodes 102 and 104 that sense physiological signals (e.g., electrocardiographic activation signals) from the heart 12. The sensed physiological signals can be used to generate electrocardiogram (ECG) data (such as ECG waveforms) that represent the sensed cardiac electrical activity over time. The ECG waveforms can represent the measured voltage amplitude over time.

[0052] As described in more detail below, certain physiological data can be transmitted (e.g., wirelessly using an antenna) to different components within the system. For example, physiological data can be transmitted from medical device 100 to receiver 106, computing system 108, and / or remote computing device 110. Receiver 106 can be a device capable of being programmed, controlled, monitored, and / or otherwise communicated with medical device 100. Receiver 106 can help facilitate communication from medical device 100 to another device or system, such as computing system 108 (e.g., laptop computer, desktop computer, server). Receiver 106 and / or computing system 108 can be communicatively coupled to another computing system 110 having a display on which a user (e.g., patient, doctor, technician) can view the data sensed and recorded by medical device 100. In some instances, receiver 106 is a mobile computing device, such as a programmer, smartphone, tablet, laptop, etc.

[0053] Medical device 100 can be programmed to initially store sensed ECG data in temporary storage, such as a buffer or cache. However, temporary storage may have limited capacity. To address this limited capacity, medical device 100 can be programmed to repeatedly delete physiological data when the amount of stored ECG data reaches the maximum capacity of the temporary storage. More specifically, the temporary storage can operate as a first-in-first-out (FIFO) buffer or a circular buffer, used to repeatedly delete the oldest data in order to store the newest data. However, under certain conditions, physiological data can be transferred and recorded in long-term storage. Some or all of the physiological data recorded in long-term storage can then be periodically transferred to another device (e.g., receiver 106, computing system 108, and / or remote computing device 110).

[0054] medical equipment

[0055] Figure 2 This is a side view of an implantable medical device 200 (hereinafter referred to as "IMD 200" for simplicity). IMD 200 can be or may resemble... Figure 1 The medical device 100 described in the text can be used in... Figure 1The system 100 is used. While the IMD200 is implantable, other types of medical devices may include the features described herein.

[0056] The IMD 200 includes an outer housing extending between a first end 202 and a second end 204. Figure 2 In the example, IMD 200 includes a first housing region 206, a second housing region 208, a third housing region 210, a fourth housing region 212, a first electrode 214, and a second electrode 216. Each of the housing regions can be a separate component, which is assembled together during manufacturing to create the outer housing of IMD 200. When assembled together, the housing regions can create a sealed outer shell. Although Figure 2 The diagram shows four separate housing sections, but an IMD 200 can be created using additional or fewer separate sections. As will be described in more detail below, one or more housing sections may comprise a ceramic material.

[0057] Figure 3 A partial exploded view of the IMD 200 is shown. The first housing region 206 includes a first cavity 218 defined by one or more inner surfaces 220 of the first housing region 206. A second electrode 216 is coupled to the first housing region 206. In some embodiments, the first housing region 206 comprises a ceramic material.

[0058] The second housing region 208 includes a second cavity 222 defined by one or more inner surfaces 224 of the second housing region 208. Figure 2 In the diagram, the second housing section 208 is shown as being composed of multiple housing components. For example, the second housing section 208 (like other housing sections) may be assembled (e.g., welded together) from multiple components to create part of the outer housing of its IMD 200. In some instances, the second housing section 208 comprises a metallic material such as titanium. In other instances, the second housing section 208 comprises a ceramic material.

[0059] The third housing region 210 may include or may be a battery assembly (which may include one or more batteries). The third housing region 210 may be formed on the exterior of the battery assembly, where one or more batteries (e.g., rechargeable battery cells) are housed. A first electrode 214 is disposed at an end of the third housing region 210. In some embodiments, the first electrode 214 is integrated with the battery assembly.

[0060] The fourth housing section 212 can serve as an interface or coupler between the first housing section 206 and the second housing section 208. For example, the fourth housing section 212 can be used to couple the first housing section 206 to the second housing section 208. More specifically, in examples where the fourth housing section 212 comprises a metal such as titanium, the fourth housing section 212 can be assembled to the second housing section 208 by welding (e.g., laser welding). And, in examples where the first housing section 206 comprises ceramic, the fourth housing section 212 can be brazed to the first housing section 206. In such examples, the fourth housing section 212 can be coupled between the first housing section 206 and the second housing section 208. Figure 3 As shown, the fourth housing region 212 can be formed as a continuous ring with an opening therethrough. Furthermore, the fourth housing region 212 may include joining features, such as one or more thinned regions or flange regions, enabling the fourth housing region 212 to be connected to other regions (e.g., by welding and / or brazing). For example, portions of other housing regions may overlap with the thinned or flange regions of the fourth housing region 212 to provide overlapping surface areas.

[0061] Figure 3 A circuit board 226 having a first circuit board region 226A and a second circuit board region 226B is shown. To illustrate the various components of the IMD 200 in an exploded view, the first circuit board region 226A and the second circuit board region 226B are shown as two separate components, but these two regions can form a single circuit board. The first circuit board region 226A and the second circuit board region 226B can include rigid circuit boards or flexible circuits (e.g., flexible circuits comprising polyimide). In the example where the first circuit board region 226A and the second circuit board region 226B are separate components, one region can include a rigid circuit board, and the other region can include a flexible circuit. Furthermore, the two regions can be electrically coupled to each other.

[0062] When the IMD 200 is assembled, the first circuit board region 226A is located within the first cavity 218, and the second circuit board region 226B is located within the second cavity 222. When the IMD 200 is assembled, a portion of either or both of the first circuit board region 226A and the second circuit board region 226B can extend through the fourth housing region 212. Therefore, in an example with a single continuous circuit board, the circuit board 226 can extend within the first cavity 218 and the second cavity 222 and through the opening in the fourth housing region 212.

[0063] Antenna 228 is located within the first cavity 218 and coupled to circuit board 226, such as to the first circuit board region 226A. As an example, antenna 228 may be embedded within the first circuit board region 226A. In this example, antenna 228 may be formed from conductive traces in the first circuit board region 226A. As another example, antenna 228 may be formed on the inner surface 220 of the first housing region 206 and electrically coupled (e.g., directly or indirectly) to conductive traces in the first circuit board region 226A. In this example, if the inner surface 220 is otherwise a ceramic material, a portion of the inner surface 220 may be metallized. As another example, antenna 228 may be embedded in the first housing region 206 and electrically coupled to conductive traces in the first circuit board region 226A. In other instances, antenna 228 may be located within an outer housing between the third housing region and the first electrode 214 (e.g., in the space between the battery and the first electrode 214). In other instances, antenna 228 may be located outside the outer housing.

[0064] Various other electrical components may be coupled to circuit board 226, such as on a second circuit board section 226B. For example, medical device 200 may include one or more memory components 230 coupled to circuit board 226. Memory 230 may include temporary memory 230A (e.g., random access memory, cache memory) and long-term memory 230B (e.g., flash memory). In some instances, temporary memory 230A and long-term memory 230B are separate sections or modules in a single chip package. In other instances, temporary memory and long-term memory are separate components attached separately to circuit board 226. In some instances, temporary memory 230A operates as memory (e.g., FIFO buffer memory), automatically deleting (or overwriting) old data when new data is stored in temporary memory 230A—while long-term memory 230B is not operated to automatically delete (or overwrite) old data, but instead manages its storage capacity by comparing the priority of physiological data already stored in memory with the priority of new physiological data. Examples of how physiological data for certain cardiac events can be recorded in long-term storage 230B may take precedence over physiological data for other cardiac events. In some instances, long-term storage 230B may have a larger storage capacity than temporary storage 230A (e.g., a larger number of bytes, kilobytes, megabytes, gigabytes, etc.).

[0065] Other electrical components may also be coupled to circuit board 226, such as one or more integrated circuits (e.g., application-specific integrated circuits, field-programmable gate arrays), which are programmed to perform processing and / or communication functions such as IMD 200. For example, electrical components like integrated circuits may include one or more processors 232 (e.g., microprocessors) coupled to memory 230 or other memory components having instructions (e.g., in the form of firmware and / or software) for performing the functions of IMD 200. Example functions of the circuitry include processing physiological data (e.g., converting sensed electrical signals from electrodes into ECG data, detecting potential cardiac events based on the sensed electrical signals, storing physiological data in memory, etc.). Electrical components may be electrically coupled to one or more batteries, such that the electrical components are powered by one or more batteries.

[0066] Figure 4 Another view of IMD 200 is shown. (As shown) Figure 4 As shown, in some instances, charging coil 234 (e.g., a set of coils comprising a conductive material) is coupled to circuit board 226, such as a first circuit board region 226A. Charging coil 234 is arranged to receive an external signal (e.g., an electromagnetic signal generated by a power transmitter) that induces a current in charging coil 234, allowing the current to recharge the batteries in the battery assembly. When the IMD 200 is assembled, charging coil 234 is located within a first cavity 218 of the first housing region 206, allowing electromagnetic signals to pass through the housing and reach charging coil 234. In some instances, charging coil 234 is wound such that a hole or space exists through the center of charging coil 234 (e.g., charging coil 234 has a ring shape or similar shape). Charging coil 234 (and other electrical components) is electrically coupled to one or more batteries 236 of the IMD 200.

[0067] Methods for PVC testing and recording

[0068] As mentioned above, to save power and storage capacity, medical devices may ultimately not record and transmit all physiological data sensed by the medical device (e.g., ECG data). More specifically, while ECG data may initially be temporarily stored in buffer or cache memory, most ECGs can be permanently deleted (rather than saved to long-term memory) as the buffer / cache memory fills with new ECG data. However, medical devices can be programmed to record and transmit certain physiological data associated with cardiac events, such as PVCs (cardiac ventricular contractions), and other types of cardiac events.

[0069] However, even limiting the recording and transmission of physiological data associated with PVCs can be an inefficient use of the power and storage capacity of medical equipment. The various methods described in this article can help determine which PVC episodes (and underlying ECG data) should be prioritized for recording and transmission.

[0070] Figure 5 An overview of the steps of method 300 for use with medical devices, such as medical devices 100 and 200 described herein, is shown. Method 300 in Figure 2-4 The following description is made in the context of medical device 200, but method 300 can also be performed by other medical devices.

[0071] Method 300 includes detecting that the first PVC occurs within a first length of ECG data ( Figure 5 (Block 302). In some instances, not all occurrences of PVCs will result in a PVC outbreak being detected or otherwise identified as ECG data with a PVC outbreak. For example, medical device 200 may be programmed with thresholds that must be met before a PVC outbreak is determined to have occurred. As an example, the threshold could be a minimum number or minimum percentage of PVCs within a given time period. In some instances, the minimum threshold is set to a minimum 5% PVC load.

[0072] The medical device 200 can be programmed to distinguish or classify different types of PVC attacks. For example, one type of PVC attack may be an attack with a maximum load value within a given time period (e.g., a time interval of about one minute or longer, such as 1-2 minutes). Another type of PVC attack may be attacks that are paired (e.g., consecutive PVCs) or triplet (e.g., triplet PVCs). Yet another type of PVC attack may be an attack with a specific morphology or shape.

[0073] Return to reference Figure 5 The first PVC outbreak detected in box 302 may be associated with one or more types of PVC outbreaks. For example, medical device 200 may be programmed to determine the maximum load value of the detected PVC outbreak, whether the detected PVC outbreak includes pairs or triplets, and / or the morphology of the PVCs present during the PVC outbreak.

[0074] In response to the detection of a first PVC, the length of the ECG data associated with the first PVC can be recorded in long-term memory. Figure 5(See block 304 in the original text). For example, ECG data may initially be stored in temporary memory, and then, in response to the detection of a PVC, a certain length of ECG data may be recorded into long-term memory. The ECG data initially stored in temporary memory may be deleted as usual when new ECG data is stored in temporary memory (e.g., in a FIFO priority manner). In some instances, the length (e.g., amount of time) of ECG data recorded into long-term memory may depend on the occurrence of the PVC and the length of ECG data still available in temporary memory. For example, medical device 200 may identify or otherwise estimate the start and end of a PVC and record ECG data between the start and end, or record a predetermined amount (e.g., 10-60 seconds) of ECG data before the start and after the end of a PVC—provided that the ECG data has not yet been deleted from temporary memory.

[0075] Method 300 also includes detecting that the second PVC occurs after the first PVC and within a second length of ECG data ( Figure 5 (Block 306 in the document). As previously stated, the medical device 200 can be programmed with thresholds that must be met before a PVC attack is determined to have occurred.

[0076] Method 300 includes determining whether the second PVC seizure is of the same type as the first PVC seizure or a different type of seizure than the first PVC seizure. Figure 5 Block 308 in the middle.

[0077] Method 300 further includes determining whether to overwrite the ECG data of the first length or to record the ECG data of the second length separately into memory. Figure 5 (Block 310 in the document). Determining whether to overwrite ECG data in long-term memory or to record ECG data associated with subsequent PVC events separately can depend on the type of the subsequent PVC event. In some instances, when a subsequent PVC event is of the same type as a previous PVC event, ECG data containing the previous PVC event can be overwritten by ECG data containing the subsequent PVC event.

[0078] For example, the maximum PVC load values ​​of two PVC events can be compared, and the PVC event with the highest load value (and its ECG data) can be recorded in long-term memory. If a previous PVC event is associated with a higher PVC load, the subsequent PVC event can be ignored. However, if a subsequent PVC event has a higher load than the PVC event already stored in long-term memory, the subsequent PVC event (and its ECG data) may overwrite the previous PVC event (and its ECG data). If a subsequent PVC event has an even higher load, the subsequent PVC event may overwrite the previous PVC event.

[0079] Alternatively, the medical device 200 may be programmed to maintain the maximum number of representative PVC episodes (e.g., ECG data from the three PVC episodes with the highest PVC load) instead of recording a single PVC episode with the highest PVC load. As another alternative, the medical device 200 may be programmed to record examples of ECG data associated with different PVC load values, such as one or more representative PVC episodes with a load value of 5-10%, one or more representative PVC episodes with a load value of 10-20%, one or more representative PVC episodes with a load value greater than 20%, and so on. Each of the load value ranges can be considered a different bucket, and a single representative PVC episode can be assigned to each bucket—provided that a representative episode has been detected for a given bucket.

[0080] Using the method described above, ECG data with the highest PVC load value are prioritized over PVC episodes with lower PVC load values. This method can save power and storage resources on medical device 200 because some PVC episodes (e.g., episodes with lower PVC load values) can be ignored and not recorded in long-term memory, thus limiting power and storage capacity consumption.

[0081] As another example of covering previous PVC attacks, a similar approach can be used for PVC attacks involving paired events and / or triplet events. Medical device 200 can be programmed to determine the number of paired and / or triplet events occurring during a PVC attack. In some instances, a minimum number of paired or triplet events must occur within a given time period before medical device 200 determines that a PVC attack has occurred. When medical device 200 detects paired and / or triplet events, it can determine whether ECG data from the same type of event has already been recorded in long-term memory, and if so, whether the current PVC attack includes the same, fewer, or more paired and / or triplet events. If the current PVC attack has more paired and / or triplet events, the ECG data of the current PVC attack can be written over the ECG data of the previous PVC attack.

[0082] In some instances, overwriting a previous PVC episode recorded in long-term memory with a subsequent PVC episode involves recording on previously recorded ECG data, deleting previously recorded ECG data before writing new ECG data, and / or recording the subsequent PVC episode separately in long-term memory, but associating the previously recorded ECG data with a flag or indicator that prevents the previously recorded ECG data from being transmitted by the medical device 200 to another device, so that the subsequent PVC episode is transmitted instead of the previous PVC episode.

[0083] An optional exception to the overwriting method described herein is that the medical device 200 may be programmed to record ECG data associated with a first PVC attack detected during a given time period to long-term memory (and prevent overwriting). More specifically, the medical device 200 may be programmed to transfer any recorded ECG data to long-term memory once daily, twice daily (e.g., daytime, nighttime), every 4–12 hours (e.g., 4 hours, 8 hours, 12 hours), once in communication with a receiver, or at certain other time intervals or triggers (e.g., in response to a command). After the recorded ECG data has been transferred to a different device, previously recorded ECG data may be deleted or otherwise marked for deletion, and the medical device 200 may be restarted by detecting and recording new PVC attacks for new time periods. The first PVC attack detected (e.g., the first PVC attack that meets a minimum threshold) may be recorded to long-term memory, and the medical device 200 may prevent the first PVC attack from being overwritten by subsequent PVC attacks, regardless of whether the subsequent PVC attacks have a higher PVC load value, occur in pairs, occur in triplets, etc.

[0084] When a subsequent PVC attack differs in type from a previous PVC attack, ECG data containing the subsequent PVC attack can be written separately to long-term memory, so that both the previous and subsequent PVC attacks are recorded in long-term memory. For example, long-term memory may separately store at least one PVC attack with paired events, another PVC attack with three consecutive events, another PVC attack with the maximum PVC load, another PVC attack representing the median PVC load, and another PVC attack as the first PVC attack detected by medical device 200. In some instances, the number of PVC attacks can be customized for each patient. For example, a medical device used with a patient having a lower average or median PVC load (e.g., ~10%) may have a lower maximum number of PVC attacks (e.g., 3 attacks) stored in long-term memory compared to a patient with a higher average or median PVC load (e.g., ~25%), while the PVC load may have a maximum number (e.g., 5 attacks). Such a maximum value can be automatically updated by medical device 200 based on historical PVC data, or it may be the result of reprogramming medical device 200.

[0085] In addition to detecting the different types of PVC outbreaks mentioned above, the medical device 200 can also be programmed to detect different forms or shapes of PVC. Figure 6An example of a short bar 400 containing ECG data of three PVCs 402, 404, and 406 is shown. PVCs 404 and 406 occur consecutively (e.g., there is no normal heartbeat between PVCs) and are examples of PVCs occurring in pairs. The shape of the PVCs can be analyzed by medical device 200 to determine whether a given PVC episode contains PVCs with similar or different shapes. If medical device 200 detects a PVC episode (e.g., by meeting a minimum threshold), and medical device 200 detects two or more PVCs with different shapes or morphologies, the ECG data associated with the PVC episode can be recorded in long-term memory.

[0086] As described above, medical device 200 can be programmed to periodically transmit any ECG data recorded in long-term storage (e.g., wirelessly) to another device, such as a receiver. The ECG data can ultimately be transmitted to a clinic for a doctor to review.

[0087] To aid in viewing, metadata can be generated and used to represent detected PVCs when bar 400 is displayed. For example, a single PVC can be represented by annotation type 408, pairs can be represented by annotation type 410, and so on. In other instances, the same annotation shape / method is used for all PVCs, but is reused for each individual PVC (e.g., pairs of PVCs will be associated with two separate annotations—one annotation per individual PVC).

[0088] In some instances, ECG data is transmitted and processed by a computing system, such as a server, before being transmitted to the clinic. Because servers have significantly more computing and storage capacity than medical devices, they can provide a second review of the ECG data before clinical examination. Servers can be programmed to filter out false PVCs and noise, confirm PVCs, group similar-shaped PVCs together, determine which PVCs require the highest priority, and / or determine confidence values ​​for individual PVCs and PVCs to prioritize certain PVCs over others, among other functions.

[0089] To support the functionality described above, the server may include one or more machine learning models (e.g., a type of deep neural network) to process ECG data for classifying cardiac activity. For example, the machine learning model may compare ECG data with labeled ECG data to determine the most closely similar labeled ECG data. The labeled ECG data may identify specific cardiac events, such as PVCs (cardiac ventricular contractions). In some instances, the machine learning model includes two paths, where the first path is a deep convolutional neural network and the second path is a deep fully connected neural network. The deep convolutional neural network receives one or more sets of heartbeats (e.g., a heartbeat string with 3-10 heartbeats), which are processed through a series of layers on the deep convolutional neural network. This series of layers may include convolutional layers for performing convolutions on the time-series data in the heartbeat string; batch normalization layers for normalizing the outputs from the convolutional layers (e.g., centering the results around the origin); and non-linear activation function layers for receiving the normalized values ​​from the batch normalization layers. The heartbeat string is then passed through a repeating set of layers, such as another convolutional layer, a batch normalization layer, and a non-linear activation function layer. This group of layers can be repeated multiple times.

[0090] A deep fully connected neural network can receive RR interval data (e.g., the time interval between adjacent heartbeats) and process the RR interval data through a series of layers: a fully connected layer, a non-linear activation function layer, another fully connected layer, another non-linear activation function layer, and a regularization layer. The outputs from both paths are then fed to the fully connected layer. The resulting values ​​are passed through the fully connected layer and a softmax layer to produce a probability distribution for heartbeat categories.

[0091] If the machine learning model determines that the ECG data most closely resembles the labeled ECG data associated with the cardiac event, then the machine learning model 108 can determine that the patient experienced the cardiac event. Furthermore, the machine learning model can measure or determine certain characteristics of the patient's cardiac activity based on the ECG data. For example, the machine learning model can determine the patient's heart rate, duration, or heartbeat count during the cardiac event based on the ECG data. The server stores the cardiac event and associated metadata (such as heartbeat classification, heart rate, duration, heartbeat count, etc.) in a database for storage.

[0092] The output of the machine learning model and the underlying ECG data can be transmitted to the clinic for doctors to view.

[0093] Computing devices and systems

[0094] Figure 7This is a block diagram depicting an illustrative computing device 500 according to an example of this disclosure. The computing device 500 may include any type of computing device suitable for implementing aspects of the disclosed subject matter. Examples of computing devices include dedicated or general-purpose computing devices, such as workstations, servers, laptops, desktops, tablets, handheld devices, smartphones, general-purpose graph processing units (GPGPUs), etc. Each of the various components shown and described in the figure may contain its own set of dedicated computing device components, such as... Figure 7 As shown and described below. For example, each of a medical device, receiver, and computing system can include its own set of components (or a subset thereof), such as Figure 7 As shown and described below.

[0095] In some cases, computing device 500 includes a bus 510 that directly and / or indirectly couples one or more of the following devices: processor 520, memory 530, input / output (I / O) port 540, I / O component 550, and power supply 560. Computing device 500 may also include any number of additional components, different components, and / or combinations of components.

[0096] Bus 510 can represent one or more buses (e.g., address bus, data bus, or a combination thereof). Similarly, in some cases, computing device 500 may include multiple processors 520, multiple memory units 530, multiple I / O ports 540, multiple I / O components 550, and / or multiple power supplies 560. Furthermore, any number or combination of these components may be distributed and / or replicated across multiple computing devices.

[0097] In some cases, memory 530 includes computer-readable media in the form of volatile and / or non-volatile memory, and may be removable, non-removable, or a combination thereof. Examples of media include random access memory (RAM); read-only memory (ROM); electrically erasable programmable read-only memory (EEPROM); flash memory; optical or holographic media; magnetic tape cassettes, magnetic tape, disk storage devices, or other magnetic storage devices; data transfer; and / or any other media that can be used to store information and is accessible by a computing device. In some cases, memory 530 stores computer-executable instructions 570 for causing processor 520 to implement aspects of instances of the components discussed herein and / or perform aspects of instances of instances of the methods and processes discussed herein. Memory 530 may include non-transitory computer-readable media storing computer-executable instructions 570.

[0098] The computer-executable instructions 570 may include, for example, computer code, machine-usable instructions, and program components, such as those executable by one or more processors 520 (e.g., microprocessors) associated with the computing device 500. The program components can be programmed using any number of different programming environments, including various languages, development kits, frameworks, and / or the like. Some or all of the functionality envisioned herein may also or alternatively be implemented in hardware and / or firmware.

[0099] According to examples, instruction 570 may be configured to be executed by processor 520 and, upon execution, cause processor 520 to perform certain processes. In some cases, processor 520, memory 530, and instruction 570 are part of a controller, such as an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), and / or the like. Such a device can be used to perform the functions and steps described herein.

[0100] I / O component 550 may include presentation components configured to present information to a user, such as display devices, speakers, printing devices, and / or similar devices, and / or input components such as microphones, joysticks, satellite dish antennas, scanners, printers, wireless devices, keyboards, pens, voice input devices, touch input devices, touch screen devices, interactive display devices, mice, and / or similar devices.

[0101] The devices and systems described herein can be coupled via network communication, which may include local area networks (LANs), wide area networks (WANs), cellular data networks, or via the Internet using an Internet service provider.

[0102] Aspects of this disclosure are described with reference to flowchart illustrations and / or block diagrams of methods, apparatus, systems, and computer program products. It should be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions.

[0103] Various modifications and additions may be made to the exemplary embodiments discussed without departing from the scope of the invention. For example, while the above embodiments relate to specific features, the scope of the invention also includes embodiments with different combinations of features and embodiments that do not include all the described features. Therefore, the scope of the invention is intended to include all such alternatives, modifications, and variations falling within the scope of the claims, as well as all their equivalents.

Claims

1. A method comprising: The first premature ventricular contraction (PVC) was detected within the first length of electrocardiogram (ECG) data, where the first PVC was a type 1 PVC. In response to the detection, ECG data of the first length is recorded in the memory; A second PVC attack was detected after the first PVC attack and within a second length of ECG data; Determine whether the second PVC outbreak is a first type of PVC outbreak or a second type of PVC outbreak; as well as The determination of whether to overwrite the first length of ECG data or to record the second length of ECG data separately in the memory is based at least in part on determining whether the second PVC outbreak is the first type of PVC outbreak or the second type of PVC outbreak.

2. The method according to claim 1, further comprising: It was determined that the second PVC outbreak was a PVC outbreak of the first type; as well as In response, the ECG data of the first length is overwritten with ECG data of the second length.

3. The method according to any one of claims 1 or 2, wherein, The first type of PVC outbreak and the second type of PVC outbreak are maximum load outbreaks, wherein the first PVC outbreak is associated with a first load value, wherein the second PVC outbreak is associated with a second load value, wherein in response to the second load value being greater than the first load value, the first length of ECG data is covered by the second length of ECG data.

4. The method according to claim 3, further comprising: A third PVC attack was detected after the second PVC attack and within a third length of ECG data, wherein the third PVC attack was associated with a third load value; as well as After determining that the third load value is greater than the second load value, the second length of ECG data is overwritten with the third length of ECG data.

5. The method according to claim 3 or 4, wherein, The first load value and the second load value are based on the number of PVCs within a predetermined time period.

6. The method according to any one of claims 1 or 2, wherein, The first type of PVC attack and the second type of PVC attack are paired events and / or triple events, wherein the first PVC attack is associated with a first number of pairs or triples, wherein the second PVC attack is associated with a second number of pairs or triples, wherein, in response to the second number of pairs or triples being greater than the first number of pairs or triples, the first length of ECG data is covered by the second length of ECG data.

7. The method according to claim 1, wherein, The first type and the second type are different types of PVC eruptions, and the method further includes: It was determined that the second PVC outbreak was a second type of PVC outbreak; and In response, ECG data of the second length is recorded in the memory, such that the memory contains both ECG data of the first length and ECG data of the second length.

8. The method according to claim 7, wherein, The first type of PVC attack is a maximum load attack, while the second type of PVC attack is a paired attack or a triple attack.

9. The method according to any one of claims 1 to 8, wherein, The detection of the first PVC outbreak is based at least in part on the satisfaction of a minimum PVC threshold.

10. The method according to any one of claims 1 to 9, further comprising: Before recording the first length of ECG data into the memory, the first length is temporarily stored in a buffer.

11. The method according to claim 10, wherein, ECG data that was not recorded in the memory was permanently deleted.

12. The method according to any one of claims 1 to 11, further comprising: Any PVC outbreaks recorded in the memory will be wirelessly transmitted to a remote computing device.

13. A computer program product comprising instructions that cause one or more processors to perform the steps of the method of claims 1-12.

14. A computer-readable medium having stored thereon a computer program product according to claim 13.

15. An implantable medical device comprising the computer-readable medium of claim 14.

Images