defibrillator

Abstract

The present application provides a defibrillator. The defibrillator is used to analyze electrocardiogram (ECG) data to identify whether it is beneficial for a caregiver to apply an electric shock to the heart in an effort to restore the heart to a normal pattern and resume a sustained strong beat. By continuously checking conditions that have been pre-verified using a comprehensive patient database as having a high predictive value (e.g., having a low false positive rate), a shockable rhythm can be identified quickly (e.g., in less than 6 seconds, in less than 3 seconds, possibly in less than 1 second) and without having to analyze the ECG data in a longer time period than would otherwise be required using conventional methods.

Description

[0001] This application is a divisional application of an invention patent application with an application date of January 13, 2017, application number 201780003386.3, and invention name “Rapid Identification of Shockable or Non-Shockable Rhythm in ECG Data”. Technical Field

[0002] This document relates to cardiac resuscitation systems and techniques, and more particularly to defibrillators that deliver a defibrillating shock to a patient's heart through appropriately placed electrodes. Background Art

[0003] Heart attacks are a common cause of death. A heart attack occurs when a portion of the heart tissue loses circulation and becomes damaged (e.g., due to a blockage in the heart's vasculature). Heart disease and other abnormalities may lead to ventricular fibrillation (VF), an abnormal heart rhythm (arrhythmia) that causes the heart to lose its ability to pump. VF occurs when normal and regular electrical impulses are replaced by irregular and rapid impulses, which causes the heart muscle to stop contracting normally and begin to quiver. Normal blood flow stops, and if normal heart contraction is not restored, organ damage or death may result within minutes. Because the victim has no perceptible warning of the impending fibrillation, death often occurs before necessary medical assistance can arrive. Other arrhythmias may include an excessively slow heart rate known as bradycardia or an excessively fast heart rate known as tachycardia. Cardiac arrest can occur in patients in which various arrhythmias of the heart, such as ventricular fibrillation, ventricular tachycardia, pulseless electrical activity (PEA), and asystole (the cessation of all electrical activity in the heart), cause the heart to provide inadequate levels of blood flow to the brain and other vital organs that support life. If these problems are not corrected quickly (usually within minutes), the rest of the body is deprived of oxygen and the person dies. Therefore, prompt care for a person experiencing ventricular fibrillation can be the key to a positive outcome for such a person.

[0004] Cardiac arrest and other heart health diseases are the leading causes of death worldwide. The purpose of various resuscitation efforts is to maintain the body's circulatory and respiratory systems during cardiac arrest in an attempt to save the patient's life. The earlier these resuscitation efforts begin, the better the patient's chance of survival. Implantable cardioverter / defibrillators (ICDs) or external defibrillators (such as manual defibrillators or automatic external defibrillators (AEDs)) have significantly improved the ability to treat these otherwise life-threatening conditions. These devices work by applying corrective electrical pulses directly to the patient's heart. An implantable or external defibrillator can be used to treat ventricular fibrillation or ventricular tachycardia, for example, by providing a therapeutic shock to the heart in an attempt to restore normal heart rhythm. In order to treat conditions such as bradycardia, an implantable or external pacing device can provide pacing stimulation to the patient's heart until the inherent electrical activity of the heart is restored. An automatic defibrillator that constantly monitors the condition of its heart and applies a shock when necessary can be implanted in people who have had problems before. Other such persons may be provided with a wearable defibrillator in the form of a vest, such as that provided by ZOLL Medical Inc. Wearable cardioverter defibrillator products, etc. External defibrillators / monitors may be used to treat other people, such as those seen in hospitals or used by emergency medical services (e.g., R or X Some people may be treated with automated external defibrillators (AEDs) of the type commonly found in airports, public gyms, schools, shopping districts, and other public places. Examples of these AEDs are from Chelmsford, Massachusetts. AED for medical companies Automated external defibrillator or AED Automatic External Defibrillator. As provided herein, the term AED generally refers to any device or system that incorporates some method or algorithm for acquiring an electrocardiogram (ECG) signal, analyzing the ECG for suitability for delivering a defibrillation shock to a patient, and delivering a defibrillation shock. Summary of the invention

[0005] The present disclosure provides a method and system for analyzing electrocardiogram (ECG) data to identify whether it is beneficial for a caregiver to apply an electric shock to the heart in an effort to return the heart to a normal pattern and to a sustained strong beat. By continuously checking conditions that have been pre-verified by a comprehensive patient database as having a high predictive value (e.g., having a low false positive rate), shockable rhythms can be identified quickly (e.g., less than 6 seconds, less than 3 seconds, between 0.1 and 3 seconds, between 0.1 and 2 seconds, possibly less than 1 second, between 0.1 and 1 second) and without having to analyze the ECG data over a longer time period than would otherwise be required using conventional methods. In some implementations, a variable length time period of an ECG signal is processed so that one or more features of the time period can be used to determine whether the ECG rhythm provided by the patient's monitored ECG signal is shockable or non-shockable using at least a threshold accuracy level (e.g., previously verified by a gold standard of patient data as being highly predictive). For example, depending on whether more processing of the ECG signal is required to make a final determination as to whether the ECG signal is producing a shockable rhythm or a non-shockable rhythm, the variable length time segment may include an initial time period that can be incrementally extended using one or more time periods (e.g., less than or equal to 3 seconds, between 0.1 and 3 seconds, less than or equal to 2.5 seconds, between 0.1 and 2.5 seconds, less than or equal to 2 seconds, between 0.1 and 2 seconds, less than or equal to 1.5 seconds, between 0.1 and 1.5 seconds, less than or equal to 1 second, between 0.1 and 1 second, less than or equal to 0.5 seconds, between 0.1 and 0.5 seconds, etc.).

[0006] In order to make a decision as quickly as possible whether the ECG signal represents a shockable or non-shockable rhythm, the defibrillation system is preferably cautious in identifying whether an interruption in CPR chest compressions has occurred. For example, the defibrillation system may employ one or more sensors (e.g., ECG electrodes, accelerometers, motion sensors) to track chest compressions, and once the defibrillation system detects a gap or interruption in chest compressions, the system may immediately or shortly thereafter begin ECG analysis. If the system cannot conclude that a shockable or non-shockable rhythm is present before chest compressions are restarted, the system may continue to detect when the next interruption in chest compressions occurs so that ECG analysis can be easily started again.

[0007] In some implementations, an initial set of rules can be applied to an initial time period of a variable length time segment, wherein various features of the ECG signal within the initial time period are evaluated. This evaluation based on the initial set of rules is used to determine whether the patient is in a shockable state or a non-shockable state, or whether more time is needed to make a decision. If the variable length time segment is extended (with an appropriate period of time), the adjusted set of rules can be applied to the adjusted time period to further evaluate the various features of the ECG signal within the adjusted time period. Then, it can be determined based on the adjusted set of rules whether the patient is in a shockable state or a non-shockable state, or whether more time is needed to make a decision. This process can be repeated until a decision is made that the patient is in a shockable state or a non-shockable state.

[0008] In some implementations, the techniques described herein may provide one or more of the following advantages.

[0009] By providing constant verification based on conditions that are pre-verified as having a low false positive rate, the time required to identify a shockable rhythm in the patient's ECG data can be shortened. This shortening of the identification time can potentially save the patient's life. The early feedback obtained can allow the caregiver to have more time to make a decision, and therefore be ready to take a specific course of action more quickly. For example, if the defibrillator requires 5 seconds to charge, and the caregiver is notified of the existence of a shockable rhythm within 1 second of the charging time, the caregiver has 4 seconds or more to prepare to apply a shock once the charging is completed. Alternatively, if the defibrillator is charged and the rhythm is determined to be shockable using the method according to the present disclosure, the shock can be applied immediately without further analysis or other forms of delay. Using one or more large databases of patient data to pre-verify conditions can help identify accurate conditions that can be used with high confidence.

[0010] Other features and advantages will be apparent from the description and drawings, and from the claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 A victim of cardiac arrest is shown being cared for using a medical device.

[0012] Figure 2 is a block diagram showing a defibrillator with an electrode package and a compression puck.

[0013] Figure 3 is a representation of the ECG signal.

[0014] Figure 4A A flow chart illustrating an example process for identifying the presence of a shockable rhythm in ECG data.

[0015] Figure 4B to Figure 4C An example schematic diagram is shown for identifying the presence of a shockable rhythm in ECG data.

[0016] Figure 4D to Figure 4E A flow chart illustrating an example process for identifying the presence of a shockable rhythm in ECG data.

[0017] Figure 4F An example of an ECG waveform is shown.

[0018] Figure 4G to Figure 4L An example of an ECG waveform associated with CPR therapy is shown.

[0019] Figure 5A to Figure 5B An example of an ECG waveform and its characteristics is shown.

[0020] Figure 6A to Figure 6C Examples of ECG waveforms satisfying each clause are shown.

[0021] Figure 7 Examples of medical devices that may be used in implementations described herein are shown.

[0022] Figure 8 Another example of a medical device that may be used in implementations described herein is shown.

[0023] Fig. 9 An example of a computing system that can be used with the techniques described herein is shown.

[0024] Like reference numbers in the various drawings represent like elements. DETAILED DESCRIPTION

[0025] The present disclosure relates to a system and method for analyzing a patient's electrocardiogram (ECG) data to provide feedback to a caregiver (e.g., a rescuer) and / or a medical device (e.g., a defibrillator) used to treat the patient regarding whether an electric shock should be administered to the patient's heart in an effort to return the heartbeat to a normal pattern. In particular, the present disclosure describes a continuous check of pre-verified conditions to have a high degree of accuracy.

[0026] The positive predictive value (PPV) was set in this paper as:

[0027] PPV = (number of true positives) / (total number of negative judgments)

[0028] Where: a "true positive" (TP) is an event where the test makes a positive prediction and the subject has a positive result according to a gold standard provided from a comprehensive patient database, and a "false positive" (FP) is an event where the test makes a positive prediction and the subject has a negative result according to the gold standard. In this case, the positive prediction is that the ECG rhythm is shockable. The gold standard is typically a review of the ECG rhythm by two or more qualified independent cardiologists during the algorithm development phase.

[0029] The negative predictive value (NPV) was set in this paper as:

[0030] NPV = (number of true negatives) / (total number of negative judgments)

[0031] Where: a "true negative" (TN) is the event that the test makes a negative prediction and the subject has a positive result according to the gold standard, and a "false negative" (FN) is the event that the test makes a negative prediction and the subject has a negative result according to the gold standard. In this case, the negative prediction is that the ECG rhythm is non-shockable.

[0032] As provided herein, a clause is an expression that defines constraints on features of an ECG waveform contained in ECG data; if the features of the analyzed ECG waveform meet the criteria of the clause, the clause is considered to be met (or satisfied). In particular, if the criteria are met, the ECG rhythm is considered to be shockable or the ECG rhythm is considered to be non-shockable, depending on the particular clause. In the description of the present invention, when only a "clause" is referenced, both shockable clauses and non-shockable clauses are generally referred to.

[0033] In some examples, if a particular clause is applied to a portion of an ECG signal that meets a threshold time length required to achieve a particular level of accuracy (e.g., a length associated with the particular clause), then the clause is a high accuracy clause to determine whether or not a shock is acceptable. Additionally, if the same clause is applied to a portion of an ECG signal that does not meet a threshold time length (e.g., a portion of an ECG signal that is less than a threshold time length), then the clause may be a normal accuracy clause. Thus, as the length of the ECG signal portion increases, the accuracy tends to increase. Thus, there will be a minimum time segment length below which the clause is only a normal accuracy clause and is not suitable as a high accuracy clause; but for time segment lengths longer than the minimum length, the clause is then considered a high accuracy clause.

[0034] A clause that is accurate enough for one duration is not always accurate enough for all durations beyond the initial clause length. Thus, for all duration clauses that the clause rule set is to use, the clause accuracy needs to be tested against a gold standard database.

[0035] For example, there may be a clause in which the minimum time segment length for determining whether an ECG rhythm is shockable or non-shockable is 1 second (referred to as a "1 second clause"). For example, there may be a clause in which the minimum time segment length for making this determination is 2 seconds (referred to as a "2 second clause"). For example, there may be a clause in which the minimum time segment length for making this determination is 3 seconds (referred to as a "3 second clause"). Depending on how the clauses are tested relative to a gold standard of a comprehensive database of patient information, the clauses may be classified according to the associated level of certainty provided by the clauses (e.g., accuracy, which may be based on a particular combined level of sensitivity and specificity, or based on PPV and / or NPV). For example, a clause that exhibits a high predictive certainty (e.g., accuracy) based on PPV and / or NPV may be classified as a high accuracy clause, while a clause that exhibits a lower predictive certainty may not be classified as a high accuracy clause. In various implementations, for a particular analysis time period (e.g., an initial time period), if a high accuracy clause is met, a decision can be made immediately whether the rhythm represented by the analyzed ECG signal is shockable or non-shockable without further analysis. In contrast, if no high accuracy clause is encountered, the time period is extended, and the rules are adjusted to generate a new high accuracy clause for the extended time period.

[0036] For example, if no high accuracy clauses are met within a particular analysis time period (eg, an initial time period such as 1 second, 2 seconds, 3 seconds, etc.), it may be determined that more time is needed to make a definitive decision to shock or not shock.

[0037] In such an example, the new or modified rule would be used over a longer time interval to achieve a sufficient level of accuracy. In some cases, the algorithm may submit a shockable or non-shockable vote over multiple time segments over an extended period of time. Thus, an initial high accuracy clause may constitute an initial rule set that requires only an initial time period (e.g., 0.1 seconds, 0.5 seconds, 1.0 seconds, 1.5 seconds, 2.0 seconds, 2.5 seconds, or 3 seconds). The adjusted rule set requires the initial time period plus one or more additional time periods (e.g., 0.1 seconds, 0.5 seconds, 1.0 seconds, 1.5 seconds, 2.0 seconds, 2.5 seconds, 3 seconds, 6 to 9 seconds).

[0038] Clauses can be developed based on findings provided by the medical community. For example, it is well known that the ventricular fibrillation (VF) rhythm, which involves uncoordinated ventricular muscle tremors that prevent the heart from pumping blood, is shockable. It is also known that the ventricular tachycardia (VT) rhythm, which involves an acceleration of the heartbeat that originates from incorrect cardiac electrical activity, is also shockable. In contrast, the cardiac arrest rhythm, which is characterized by a flat line in which the heart is in a state of no electrical activity, is non-shockable. An ECG signal showing a clear QRS rhythm is also non-shockable.

[0039] In general, there is a continuum from coarse VF to fine VF to asystole. The boundary between fine VF and asystole can be set, for example, at an average amplitude of about 0.08-0.2 mV. On the other hand, a shockable VT rhythm can be characterized by a heart rate greater than about 160-180 bpm, and can vary from patient to patient. Generally, the higher the VT rate, the more likely it is that the patient should be shocked. Thus, the criteria for assessing the presence of a VT rhythm may need to be combined with one or more other features of the ECG waveform (e.g., morphology).

[0040] During ECG analysis, detection of regular beats representing a QRS wave will indicate that the rhythm is non-shockable. Thus, detection of a large amplitude (R wave) of short duration (e.g., a narrow wave less than 160nsec and a sharp slope greater than 20mV / se) can represent a QRS waveform. In addition, a normal sinus rhythm (non-shockable) typically has a high isoelectric content characterized by a large portion of the signal being located within a relatively narrow isoelectric band. In contrast, VF or VT signals typically have a fairly low isoelectric content. Thus, for various implementations, the clauses used for ECG analysis can be based on the above description.

[0041] The determination of whether a segment is shockable or non-shockable can be made by periodically checking (e.g., substantially continuously) whether the data collected up to the point in time when the check is made conforms to the criteria of a shockable or non-shockable clause defined for one or more parameters calculated from the ECG data. Once a clause with high NPV and PPV (also referred to as a high accuracy clause) is satisfied, the presence of a shockable or non-shockable rhythm can be identified and an output provided, e.g., appropriate feedback can be provided to a caregiver and / or a signal can be output to a medical device for treating the patient accordingly.

[0042] In order to start the analysis of ECG signals to determine the shockability of the rhythm, a short initial segment length (e.g., initial time period) is selected to start the analysis. In some implementations, this length is 1 second, but the initial analysis segment length can be shorter (e.g., as short as 100 milliseconds, 500 milliseconds, etc.).

[0043] In this way, a very fast and accurate judgment of the condition of the ECG rhythm can be achieved in a time that is an order of magnitude faster than existing traditional commercial AED analysis algorithms. The speed of judging the condition of the ECG rhythm is an important determinant of the survival of victims of cardiac arrest. Using traditional technology, the duration of the ECG segment required for proper evaluation is 9 to 15 seconds, while using various implementations of the technology disclosed in the present invention, a highly accurate judgment of the ECG rhythm can be achieved in a time as short as 1 second. This short analysis duration enables automatic ECG rhythm assessment during very short pauses in chest compressions (such as immediately before delivering a defibrillation shock or during a pause in chest compressions that are providing ventilation). Thus, no additional pauses in chest compressions are required to judge the ECG rhythm condition. It is known that pauses in chest compressions can reduce the survivability of patients with cardiac arrest.

[0044] Clauses can be defined for specific segment lengths. For example, as described in the previous paragraph, if the criteria for the shockable or non-shockable clause for the specific segment length are met, appropriate feedback can be provided to the caregiver and / or a medical device for treating the patient accordingly. However, if the shockable or non-shockable clause for the specific segment length is not met, the system determines to extend the variable length time segment using one or more additional time periods. The additional time period can have a length different from the initial time segment. For example, the first segment can be 500 milliseconds or 1 second, and further increments can be increments shorter than the initial segment, for example, 1 millisecond, 10 milliseconds, 20 milliseconds, 100 milliseconds, 250 milliseconds.

[0045] For example, if the delta duration is 20 milliseconds, then using the same initial 500 milliseconds of ECG data and the new 20 milliseconds of data, the new analysis duration may be 520 milliseconds. In this case, at least one of the criteria of the clause for the original 500 milliseconds of data may be modified, or a new clause may be added for the new length of the analysis segment (e.g., 520 milliseconds).

[0046] Optionally, the additional time period can have a duration substantially the same as the initial time period (e.g., 500 milliseconds). As previously described, one or more features of new values ​​for the variable length time segments can be identified based on an adjusted rule set corresponding to the adjusted new time period to update the determination of whether the patient is in a shockable state or a non-shockable state.

[0047] In this way, a minimal amount of time will be spent to accurately assess the ECG rhythm condition for each specific rhythm subclass.

[0048] Since periodic checks can be performed at a high degree of time granularity (e.g., 1 millisecond, 10 milliseconds, 20 milliseconds, 100 milliseconds, 250 milliseconds) or in a substantially continuous manner, for some instances, a shockable rhythm or pattern can be identified quickly (e.g., faster than conventional analysis methods) without having to wait for up to 15 seconds of ECG data. As further provided herein, for various implementations, the ECG signal can be processed according to time segments of variable lengths, wherein the initial time period of the analysis can be extended in small time increments if a specific accuracy threshold or prediction level tested against a comprehensive patient database is required to be achieved to determine whether the patient is in a shockable or non-shockable state.

[0049] In the prior art of ECG algorithms for assessing shockable / non-shockable rhythm conditions (e.g., ECG analysis related to whether a patient should be treated with a defibrillation shock), the system focuses on sensitivity and specificity and achieves high values ​​of both sensitivity and specificity at the end of a fixed time interval. For example, a typical AED ECG analysis interval is fixed to last 3 seconds, where at least two consecutive intervals are required before a shock-no-shock decision is made. Sensitivity can be characterized as the percentage of people who should be given a defibrillation shock and are correctly identified as such. Specificity can be characterized as the percentage of people who should not be given a defibrillation shock and are correctly identified. In the prior art, at least 6 seconds of data are collected so that both sensitivity and specificity exceed 90%. Typically, specificity is required to be much greater than 95%, while sensitivity is required to be greater than 92-95%. This desired combination of high sensitivity and high specificity can be described using the area under the curve (AUC) of the receiver-operator curve (ROC) well known to those skilled in the art, where both values ​​must be high in order to achieve a high ROC AUC.

[0050] The sensitivity or true positive rate (TPR) is given by:

[0051] TPR = TP / P = TP / (TP + FN)

[0052] The specificity (SPC) or true negative rate (TNR) is given by:

[0053] SPC=TN / N=TN / (TN+FP)

[0054] As can be seen in the above equation, an increase in sensitivity may come at the expense of increased false positives, while an increase in specificity may come at the expense of false negatives, so accuracy has traditionally required a combined measure of both sensitivity and specificity, resulting in longer analysis times (more than 6 to 9 seconds).

[0055] In contrast, the implementation of the present disclosure focuses on accuracy, as exemplified by NPV and PPV, as more relevant metrics for system output, thereby greatly accelerating the determination of ECG rhythm conditions. As can be seen from the equations for PPV and NPV below, the variable for PPV is related to positive prediction (e.g., "shock recommended"), and the variable for NPV is related to negative prediction (e.g., "shock not recommended"). Thus, if a positive decision results in a high PPV, it can be considered a "good decision", and if a negative decision results in a high NPV, it can also be considered a "good decision". This is not the case for sensitivity and specificity. In the implementation presented herein, negative decisions are made using the minimum amount of ECG data segment length required to achieve a high NPV or other metrics that measure the accuracy of negative predictions similar to NPV; and positive decisions are made using the minimum amount of ECG data segment length required to achieve a high PPV or other metrics that measure the accuracy of positive predictions similar to PPV.

[0056] The positive predictive value is given by:

[0057] PPV=TP / (TP+FP)

[0058] The negative predictive value is given by:

[0059] NPV=TN / (TN+FN)

[0060] The accuracy (ACC) is given by:

[0061] ACC=(TN+TP) / (TP+FP+FN+TN)

[0062] Clauses with a sensitivity greater than a threshold are sometimes referred to as high sensitivity clauses. Clauses with a specificity greater than a threshold are sometimes referred to as high specificity clauses. High accuracy is accuracy with a high NPV or PPV: In terms of sensitivity and specificity, a high accuracy clause requires both high sensitivity and high specificity. Due to the above equations for accuracy, sensitivity, and specificity, a clause that is limited to high sensitivity or high specificity is not necessarily a high accuracy clause. When determining accuracy, as known to those skilled in the art, the area under the curve (AUC) of the associated receiver operating curve (ROC) combines sensitivity and specificity into a single measure of the discriminatory power of the detection algorithm.

[0063] Although techniques using voting schemes based on analysis segments of predetermined lengths as described in more detail below may require at least 6 seconds to determine whether a patient is in a shockable state, in contrast, in some cases, a rapid analysis technique based on high-accuracy clauses according to the present disclosure can be used to make a determination in less than the typical 6 seconds of conventional voting-only techniques. In emergency resuscitation situations where administering treatment sooner rather than later may be critical to patient survival, the rapid analysis techniques provided herein can shorten potential life-saving time delays. Multiple high-accuracy clauses can be identified by pre-verifying candidate clauses on a large database of patient data, thereby enabling the identified clauses to be used together with high confidence in the rapid identification of shockable rhythms.

[0064] However, it should be understood that in cases where a judgment cannot be made based on the evaluation of clauses alone, a voting scheme can still be used. In some implementations, a voting scheme can be used in conjunction with the techniques provided herein based on high accuracy clauses. For example, if a clause is satisfied within a short period of time (e.g., half a second), analysis can continue until another clause is satisfied. This can, for example, allow the use of relatively low sensitivity clauses without compromising the accuracy of the judgment.

[0065] As described in more detail below, these techniques can be used to determine whether a patient is in a shockable or non-shockable state based on ECG time segments of variable length. An ECG time segment is a portion of an ECG signal. Compared with fixed-length techniques that select the length of a time signal for determination before analysis, in the case of using variable-length techniques, the length of the signal for determination is not known or predetermined before analysis. Instead, multiple time segments (at least some of which are different in length from other time segments) can be appropriately used to complete the analysis to make a shockable decision as quickly as possible. In addition, the length of a segment can depend on which clauses are used. Depending on, for example, which features of the signal are used to evaluate a particular clause, and / or based on the value of a parameter used to evaluate a clause, some clauses can use shorter or longer time segments than other clauses. In this way, the length of the time segment being processed can be changed based on the identification of the feature being used to determine.

[0066] In a typical scenario, a caregiver (sometimes called a rescuer) applies the electrodes of a defibrillator to a patient. The defibrillator then collects ECG data, sometimes simultaneously with CPR treatment applied to the patient by the caregiver, or sometimes at the completion of a CPR treatment cycle or between CPR treatment cycles. The defibrillator collects and processes ECG data by evaluating clauses for time segments of the collected data. If at least one high-confidence clause is met, the defibrillator uses the state represented by the clause (e.g., shockable or non-shockable) to guide the caregiver or the defibrillator itself to apply a shock. Sometimes, based on the results of applying the clauses to the time segments of the data, the shockable state can be identified within less than 6 seconds of the patient entering the shockable state, and sometimes within 2 to 3 seconds, within 1 second or less (other times a three-step voting scheme using a fixed-length time segment predetermined prior to analysis is used, where the three-step voting scheme may require at least 6 to 9 seconds to identify the patient's current state.)

[0067] In some examples, a caregiver may pause CPR therapy (e.g., due to an explicit indication to pause CPR therapy during the natural course of repetitive CPR therapy and / or during ventilation) while a portion of the data following completion of CPR therapy is being collected and analyzed for further confirmation of an initial determination as to whether the rhythm is shockable or non-shockable. Confirmation of an initial determination of shockability is sometimes referred to as a reconfirmation mode (as further described below), which may allow CPR artifacts in the signal to be filtered, eliminated, or reduced during analysis, but may be potentially dangerous to the patient depending on how long the pause in CPR therapy was. Thus, if a clause evaluation for a time segment of the data successfully determines the patient's status after a relatively short amount of time (e.g., less than 6 seconds), CPR therapy may be resumed relatively quickly, thereby reducing risk to the patient.

[0068] To provide further context, the heart relies on an organized sequence of electrical pulses to beat effectively. Deviations from this normal sequence are known as "arrhythmias". Specific medical devices include signal processing software for analyzing electrocardiogram (ECG) signals obtained from medical patients (e.g., victims at the emergency scene) to determine when there are arrhythmias such as ventricular fibrillation (VF) or shockable ventricular tachycardia (VT). These devices include automatic external defibrillators (AEDs), ECG rhythm classifiers, and ventricular arrhythmia detectors. AEDs are defibrillators--devices that deliver controlled shocks to patients--while being equally easy to use (e.g., by providing verbal prompts to care providers via assessing patient treatment to "tell" providers to attach patients to AED treatment and activate AED treatment). Medical devices can also be configured to distinguish between a variety of cardiac waveforms such as VT and VF.

[0069] As discussed herein, VT is a tachyarrhythmia caused by a ventricular ectopic focus, characterized by a rate generally greater than 120 beats per minute and having a wide QRS complex. VT can be monomorphic (usually a regular rhythm with the same QRS complex originating from a single focus) or polymorphic (unstable, an irregular rhythm with varying QRS complexes). Depending on the rate and duration of VT, a heart in a VT state may or may not produce a pulse (e.g., the pulsating movement of blood through the circulatory system). The cardiac activity in a VT state still has some organization. If there is no pulse associated with the VT rhythm, VT is considered to be an unstable and life-threatening condition. Unstable VT can be treated with electric shock or defibrillation.

[0070] Supraventricular tachycardia (SVT) is a rapid heartbeat that begins above the lower chambers of the heart (ventricles). SVT is an abnormally fast heart rate that begins in one of the upper chambers of the heart (atria), in a component of the heart's electrical conduction system called the atrioventricular (AV) node, or both. Although SVT is rarely life-threatening, its symptoms, which include a racing heart, fluttering or pounding in the chest or a feeling of increased heart rate (palpitations), or dizziness, can be uncomfortable.

[0071] VF is often an immediate life threat. VF is a pulseless arrhythmia with irregular and chaotic electrical activity and ventricular contraction, in which the heart immediately loses its ability to pump. VF is the leading cause of sudden cardiac death (SCD). The VF waveform has no associated pulse. The corrective measure for this rhythm is to defibrillate the heart using an electrical charge.

[0072] The normal heartbeat wave starts from the sinoatrial node (SA node) and moves toward the far lower corner of the left ventricle. A huge electric shock to the heart can correct VF rhythm and unstable VT rhythm. This huge electric shock can force all the heart cells in the heart to depolarize at the same time. Subsequently, all heart cells enter a short rest period. It is hoped that the sinoatrial node (SA node) will recover from the shock before any other cells, and the rhythm so produced will be a rhythm that produces a pulse (even if it is not a normal sinus rhythm).

[0073] Many AEDs implement algorithms for identifying VT waveforms and VF waveforms by performing ECG analysis at specific times during events in which patients are rescued using defibrillation and cardiopulmonary resuscitation (CPR). The first ECG analysis can be initiated within seconds after the defibrillation electrodes are attached to the patient or after a CPR treatment cycle has been administered. Based on the results of the first analysis, subsequent ECG analysis may or may not be initiated. Typically, if the first analysis detects a shockable rhythm, the rescuer is advised to deliver a defibrillation shock. After delivering the shock, a second analysis may be automatically initiated to determine whether the defibrillation treatment is successful (e.g., whether the shockable ECG rhythm is converted to a normal rhythm or other non-shockable rhythm). If the second analysis detects the continued presence of a shockable arrhythmia, the AED advises the user to deliver a second defibrillation treatment. Then, a third (and possibly fourth, fifth, etc.) ECG analysis may be performed to identify whether the shockable rhythm still exists, and the rescuer is advised to deliver additional defibrillation treatment accordingly.

[0074] Reference now Figure 1 , showing an AED 100 that can be used to deliver a defibrillation shock to a victim or patient at the appropriate time. Figure 1 A rescuer is described using an AED 100 to automatically monitor a victim during cardiac resuscitation. The AED 100 uses the measured ECG signal to monitor the victim's heart and charges the defibrillator within the AED. In some implementations, the AED may be configured to provide feedback to the rescuer (e.g., using a display device or a speaker) in the presence of a shockable rhythm. Using the techniques described herein, a determination of whether a shockable rhythm exists can be made significantly faster than the charging process of the defibrillator, allowing the rescuer to be ready to decide to administer a shock when the defibrillator is ready.

[0075] In some implementations, if the AED 100 detects a dangerous cardiac rhythm, the AED 100 generates an alarm signal for notifying the rescuer of the condition. In the event that the rescuer issues a command to the AED 100 indicating a defibrillation shock, the AED 100 may generate the shock to the victim. The defibrillation shock is intended to remedy the dangerous rhythm of the victim's heart. In this article, a cardiac rhythm that can be remedied via such a defibrillation shock is referred to as a shockable rhythm.

[0076] AED 100 also includes a charging module configured to charge the AED. In some implementations, AED 100 can be adaptively charged based on the monitored ECG signal. For example, the defibrillator can be pre-charged only when there is a shockable rhythm as determined by analyzing the monitored ECG signal. In another example, the charging level of the device can be determined and set based on the monitored ECG signal. In some implementations, the charging method (e.g., charging rate, charging time related to the resuscitation process) can be changed based on the monitored ECG signal in an effort to save power. For example, if time permits, the capacitor can be charged in a slower manner than normal charging to save power, but it is still ensured that the capacitor is fully charged when the rescuer needs the defibrillator. The charging method can be determined using the early identification of shockable rhythms that can be promoted by using the technology described herein. For example, a fast charging process can be triggered when a shockable rhythm is identified, thereby further reducing possible life-threatening delays when applying a shock. On the other hand, if a shockable rhythm is not detected, the AED 100 can be charged more slowly to save power. Alternatively, if it appears that the ECG rhythm may be shockable, but has not yet been confirmed, the defibrillator can preemptively begin charging in case the rhythm does become shockable. This feature can be useful in cases where ECG analysis occurs during chest compressions or in a reconfirmation mode (wherein the analysis of a short period of time after chest compressions is used to confirm the presence of a shockable rhythm that has been detected during chest compressions). In this case, the defibrillator can begin charging during the administration of chest compressions so that once the presence of a shockable rhythm is confirmed, the defibrillator is easily able to deliver a shock.

[0077] The AED 100 may be configured to use a rhythm suggestion method for one or more of the following: a) quantifying various features of the ECG signal (e.g., frequency domain features); b) distinguishing between normal ECG rhythms and abnormal ECG rhythms such as VF; c) detecting the onset of abnormal ECG rhythms; and d) making decisions related to the physiological state of the heart. The analog ECG signal may be converted into data that can be analyzed using one or more processors. Such data may be referred to as "ECG data," where the term is used interchangeably with the term "ECG signal" herein. Once the AED 100 identifies the presence of a shockable rhythm based on the ECG data, the display (or another output device) of the AED 100 may be used to communicate an appropriate message to the rescuer.

[0078] In some examples, one or more therapy delivery devices 110 can be connected to the AED 100. The therapy delivery device 110 can include, for example, a portable chest compression device, a drug infusion device, a ventilation device, and / or a device that can administer multiple therapies such as defibrillation, chest compressions, ventilation, and drug infusion. The therapy delivery device 110 can be physically separate from the defibrillator AED 100 and controlled via one or more communication links 120. The communication link 120 can be a wired link (e.g., a cable) or a wireless link (e.g., Bluetooth). or Wi-Fi link).

[0079] In some implementations, at least a portion of the control and coordination of the overall resuscitation process and / or the delivery of various treatments may be performed in conjunction with a device 130 or processing element located external to the AED 100. For example, the device 130 may download and process ECG data from the AED 100, for example to enable rapid detection of shockable rhythms as described herein. The device 130 may also be configured to analyze the ECG signal, make relevant determinations such as those described above and below based on the analysis, and control other treatment devices 110 including the AED 100. In other implementations, the AED 100 may perform all processing of the ECG, including analyzing the ECG signal, and may send a final determination of appropriate treatment to the control device 130, which then performs control actions on other linked devices 110.

[0080] Reference now Figure 2 , schematic block diagram 200 shows an example of a defibrillator 100, wherein the defibrillator 100 includes a set of electrodes 206 and a compression scaler 209 (e.g., for measuring the depth of chest compressions during CPR therapy). In general, the defibrillator 100 (and optionally one or more of the electrodes 206 and the compression scaler 209) defines a device for administering care to a patient, subject, or individual in need of cardiac assistance (e.g., a victim).

[0081] The AED 100 includes a switch 202 and at least one capacitor 204 for selectively supplying or applying an electric shock to a subject. In addition, the defibrillator may also include an ECG analyzer module 205, a transthoracic impedance (TTI) module 208, and a CPR feedback module 210. The CPR feedback module 210 may be used, for example, to control the frequency and amplitude of chest compressions applied to the subject, and the TTI module 208 may be used to analyze the patient's transthoracic impedance (e.g., as received via the measured TTI signal). The AED 100 also includes input / output modules such as one or more speakers 214 and one or more displays 216.

[0082] The AED 100 also includes a patient treatment (PT) module 212. The PT module 212 may include one or more processors and may be configured to receive input from one or more of the ECG analyzer module 205, the TTI module 208, and the CPR feedback module 210. The PT module 212 may use input from at least the ECG analyzer module 205 to perform various functions, such as determining whether a shockable rhythm is present in the analyzed ECG data. Figure 2 The ECG analyzer module 205 is shown as an entity separate from the PT module 212, but the two modules can be implemented as a single module. The PT module 212 can be configured to access a memory device (e.g., a hard drive or random access memory or flash memory or another memory device) to access the representation of one or more clauses, thereby determining the presence of a shockable rhythm. The PT module 212 can also include a buffer 213, which is used to accumulate or otherwise maintain the incoming ECG data accessed by the PT module 212 (or ECG analyzer module 205) to calculate one or more parameters of the accumulated data. In some implementations, the buffer 213 is a component separate from the PT module, or the buffer 213 is integrated with one of the other components described herein (such as the ECG analyzer module 205, etc.). Then, the PT module 212 can determine whether the high accuracy statement defined using one or more parameters is satisfied, thereby determining whether there is a shockable rhythm. When a shockable rhythm is determined to be present, PT module 212 may send a control signal to one or more output devices (eg, speaker 214 and / or display 216) to cause an alarm to be presented on the corresponding output device.

[0083] In this example, the ECG analyzer module 205, the TTI module 208, the CPR feedback module 210, and the patient treatment (PT) module 212 can be grouped together as a logic module 218 that can be implemented by one or more computer processors. For example, the elements of the logic module 218 can be implemented as: (i) a sequence of computer-implemented instructions executed on at least one computer processor of the defibrillator 100; and (ii) interconnected logic or hardware modules within the defibrillator 100.

[0084] exist Figure 2In the example of, electrode package 206 is connected to switch 202 via a port on defibrillator 100 so that different packages can be connected at different times. Electrode package can also be connected to ECG analyzer module 205 and TTI module 208 via the port. In this example, compression calibrator 209 is connected to CPR feedback module 210. In one implementation, ECG analyzer module 205 is a component for receiving ECG signals. Similarly, TTI module 208 is a component for receiving signals representing transthoracic impedance. Other implementations are also possible. For example, the functions of modules 205, 208 and 210 can be combined or distributed in different ways. In addition, the signals received by these modules can be distributed in different ways.

[0085] In some implementations, the PT module 212 (possibly in conjunction with other portions of the logic module 218 (eg, the ECG analyzer module 205)) can be configured to process an incoming ECG waveform to determine the presence or absence of a shockable rhythm. Figure 3 An example of an ECG waveform 250 representing ECG data that can be processed using the ECG analyzer module 205 and the PT module 212 is shown. In some examples, the presence or absence of a shockable rhythm is determined based on characteristics of a time segment of the ECG waveform 250. For example, in Figure 3 In the example, three time segments 252, 254, 256 with different lengths, each representing a portion of about 3 seconds of ECG data, can be used. One of these time segments 252 can be used to determine, for example, based on the characteristics of the time segment 252 corresponding to the value of the parameter used to evaluate the clause to be in compliance with the clause. A different time segment 254 can be used to determine that it is in compliance with the clause, wherein the clause can be the same clause as the clause applied to the first time segment 252 or a different time segment (e.g., overlapping or completely separated time segments). If the evaluation of the clauses for the three time segments indicates the presence of a shockable rhythm, an alarm can be issued to the caregiver and / or medical device. Although three time segments are shown in this example, other examples can use only one time segment, or two time segments, or other number of time segments to make a judgment at a specific time.

[0086] Figure 4A Processing flow diagram 300 shows an example process for identifying the presence of a shockable rhythm in ECG data. The process may be performed, for example, by Figure 2 The logic module 218 shown is implemented and applied to the use of Figure 3 The ECG waveform 250 represents data such as ECG data.

[0087] Process 300 includes acquiring and storing 302 ECG data in a memory buffer. The memory buffer may be configured to store ECG data based on the electrodes 206 ( Figure 2) to store or hold data using ECG signals collected by the electrode 206. For example, the ECG data collected by the electrode 206 can be quantized and digitized (e.g., using an analog-to-digital converter [ADC]) and stored in a memory buffer. The memory buffer can be configured to store multiple ECG segments based on a first-in, first-out (FIFO) basis. For example, the ECG signal can be analyzed as it enters the buffer. In some examples, approximately 2 to 3 seconds of stored ECG signal data can be used to analyze the signal.

[0088] In some implementations, where ECG analysis occurs during CPR therapy and during a brief period of time after CPR therapy, a portion or all of the ECG data being acquired is stored during a reconfirmation mode 301 of CPR therapy. In the reconfirmation mode, CPR therapy can be temporarily suspended so that the CPR therapy does not affect the signals being acquired for the purpose of confirming a recommendation whether to administer a defibrillation shock. In some examples, the reconfirmation mode is terminated and the caregiver is prompted to restart CPR therapy if sufficient ECG data has been acquired.

[0089] Processing 300 may include processing 304 at least a portion of the ECG data stored in the buffer. This may include processing the data corresponding to a time segment of the ECG signal so that one or more features in the identified time segment can be used to determine whether the patient is in a shockable state at least with a threshold level of accuracy. In some implementations, this may include accessing a portion of the data stored in the buffer and calculating one or more parameters based on the accessed portion of the data. The data is then used in a table of clauses or other database to determine whether a shockable rhythm (or non-shockable rhythm) can be identified based on the accessed portion of the data. The length of the processed time segment will vary depending on the parameters evaluated in the particular instance of processing.

[0090] The process 300 also includes accessing 306 a database of clauses defined according to one or more parameters calculated based on the accessed data portion. The clauses are defined such that various combinations of the one or more parameters can be used to identify the presence of a particular class of ECG waveforms representing a shockable or non-shockable state within a corresponding portion of the ECG signal with a corresponding level of accuracy.

[0091] For some implementations, the length of the time segment of the ECG signal depends on which clause is being applied to the time segment. For example, the clause can be stored in association with the minimum time required to achieve a threshold accuracy (e.g., 99%). In other words, the length of the time segment of the ECG signal can be changed according to the clause applied. As an example, one clause may require 1 second of ECG data to achieve a 99% accuracy value, while another clause may require 3 seconds of ECG data to achieve a 99% accuracy value. In this way, since each clause is associated with a specific parameter, the length of the time of the ECG signal can depend on the identification of the parameter being evaluated at a given time. Processing 300 may include: retrieving the time length associated with a specific clause (e.g., retrieving the value of the time length stored in the data storage device), and determining whether there is enough ECG data available for applying a specific clause with a threshold accuracy value. However, it is understood that for a specific implementation, a given clause can be used to achieve a threshold accuracy level for multiple time periods analyzed for ECG.

[0092] In some implementations, the accessed clauses can be sorted based on their specificity. For example, the clauses corresponding to a given set of parameters can be sorted so that the parameters are tested for the high accuracy clauses before the clauses with relatively low accuracy are tested. Due to the high confidence level associated with the high accuracy clauses (due to the corresponding low false positive rate), if the shockable rhythm is detected based on the high PPV clause, it may not be necessary to test other clauses with low PPV, thereby shortening the overall time required for the existence of the shockable rhythm. If the clause is not satisfied within the specific analysis duration, it may be necessary to process the data corresponding to the additional segments of the ECG data before making the final decision related to the presence or absence of shockability. That is, more time can be allocated to the ECG analysis to make a decision. For example, one or more incremental time amounts can be added to the initial time period of the ECG analysis to make the appropriate decision whether to apply a defibrillation shock. As discussed herein, the incremental time amount can depend on multiple factors, such as the type of the clauses adopted, the ECG signal mode, the ECG analysis history, etc. Any suitable incremental time amount may be added to the variable length time segments of the ECG analysis, for example, less than about 3 seconds (e.g., between 1 and 3 seconds, between 0.1 and 3 seconds), less than about 2.5 seconds, less than about 2 seconds (e.g., approximately 2 seconds, between 1 and 2 seconds), less than about 1.5 seconds, less than about 1 second (e.g., approximately 1 second, between 0.1 and 1 second), less than about 0.5 seconds (e.g., approximately 0.5 seconds, between 0.1 and 0.5 seconds), less than about 0.1 second, etc.

[0093] Therefore, in the case where it is determined that the total time of the analysis should be extended, the additional time increment may be different from the initial time period and / or other increments. Thus, the additional time increment may be shorter or longer than the initial time period of the ECG analysis. For example, the initial time period of the analysis may be 0.1 to 3 seconds (e.g., 0.1 seconds, 0.5 seconds, 1 second, 1.5 seconds, 2 seconds, 2.5 seconds, 3 seconds, etc.), but the time increment extension for further analysis may be 0.1 to 0.5 seconds, 0.5 to 1 second, 1 to 2 seconds, etc. In addition, the subsequent time increment extensions of the analysis may be different. As an example, the first time increment extension used for the analysis may be 0.1 to 0.5 seconds, and the subsequent time increment extension used for the analysis may be 1 to 2 seconds. Other combinations of time periods and incremental extensions are also possible. It can also be understood that for specific situations, the additional time increment may be substantially the same as the initial time period.

[0094] Any suitable initial time period for ECG analysis of variable length time segments can be used. For example, the initial time period for analysis can be between about 0.1 to 3.0 seconds, between about 0.5 to 1.5 seconds (e.g., approximately 1 second), between about 1.5 to 2.5 seconds (e.g., approximately 2 seconds), between about 2.5 to 3.5 seconds (e.g., approximately 3 seconds), or any other suitable time period.

[0095] One or more clauses can constitute a rule set corresponding to a specific duration of ECG analysis to determine whether the patient is in a shockable state or a non-shockable state. As described herein, clauses and / or rule sets can be statistically determined to correspond to a specific level of accuracy. An initial rule set (comprising a set of clauses, such as high accuracy clauses, etc.) can be applied to the initial time period of ECG analysis for a time segment of variable length, wherein: various features of the ECG signal in the initial time period are evaluated to determine whether the rhythm is shockable or non-shockable, or to determine whether the time segment of the variable length should be extended to make an appropriate decision. By extending the time segment of the variable length (e.g., in an incremental manner), an adjusted rule set (comprising an adjusted set of clauses, such as a clause requiring a longer duration than the initial high accuracy clause) can be applied to the adjusted time period to further evaluate the various features of the ECG signal in the adjusted time period, thereby determining whether the rhythm is shockable or non-shockable, or to determine whether more time is needed to make a decision. The process can be repeated (e.g., iteratively) until an appropriate decision is made regarding whether the patient is in a shockable state or a non-shockable state.

[0096] In some implementations, a voting scheme may be employed to determine the presence or absence of shockability using clauses that do not have high accuracy. The voting scheme uses fixed-length time segments rather than variable-length time segments (e.g., time segments that can be changed depending on which of the multiple clauses are applied to the data, or that can be divided into multiple voting segments). For example, data corresponding to three separate segments of ECG data may be processed to mark the segments as shockable or non-shockable, and the final decision may be based on a mark corresponding to at least two of the three marks. If the first two segments are marked as shockable, the voting scheme may terminate and the presence of a shockable rhythm may be identified. If the first segment is marked as shockable and the second segment is marked as non-shockable, the third segment is evaluated. In such a voting scheme, each segment is typically a fixed length, such as 3 seconds. Thus, using only such a voting scheme, a minimum amount of time has passed before a determination can be made as to whether the patient is in a shockable state or in a non-shockable state. For example, if 3 second segments are used, at least 6 seconds and at most 9 seconds elapse before a decision can be made.

[0097] In contrast, in some implementations, delays inherent in voting schemes can be avoided, for example, by using high accuracy clauses to determine the presence or absence of a shockable rhythm. High accuracy (e.g., low false positive rate) clauses can be created or defined in various ways. Use of such high accuracy clauses can allow the presence or absence of a shockable rhythm to be identified within a short time window (e.g., less than 1 second), thereby shortening the analysis time compared to, for example, the analysis time associated with a voting scheme. In some implementations, high accuracy clauses are heuristically determined by testing the accuracy of various candidate clauses against a database of pre-stored patient data to determine clauses with low false positive rates. In some implementations, clauses can be determined, for example, by using machine learning processing on the database to identify conditions that indicate the presence of a shockable rhythm with a low false positive rate.

[0098] Although using high accuracy clauses is often faster than using a voting scheme, a voting scheme may still be used in certain situations, such as situations where no high accuracy clause is met and thus the high accuracy clause cannot be used to determine whether the patient is in a shockable state or a non-shockable state. In some implementations, for example, a voting model may be employed in parallel with other (faster) analysis techniques described herein.

[0099] Clauses with a sufficient level of accuracy for a specific time segment length (hereinafter referred to as "high accuracy" clauses) and clauses with an insufficient level of accuracy for a specific time segment length (also referred to herein as "normal accuracy") are defined using parameters calculated by processing ECG data stored in a memory buffer.

[0100] In some examples, a high accuracy clause is defined as having a 99% accuracy threshold for a particular time length (width) of a waveform. In other words, a clause is a high accuracy clause if it is associated with a time length of 3 seconds and has at least 99% accuracy.

[0101] In some examples, if a particular clause is applied to a portion of an ECG signal that meets a threshold time length required to achieve a particular level of accuracy (e.g., a length associated with the particular clause), then the clause is a high accuracy clause. Additionally, if the same clause is applied to a portion of an ECG signal that does not meet a threshold time length (e.g., a portion of the ECG signal whose length is less than the threshold), then the clause may be a normal accuracy clause. Thus, as the length of the ECG signal portion increases, the accuracy tends to increase. Thus, there will be a minimum time segment length below which a clause is only a normal accuracy clause and is not suitable as a high accuracy clause, while for time segment lengths longer than the minimum time length, the clause is a high accuracy clause.

[0102] For example, there may be a clause where the minimum time segment length is 1 second (referred to as a "1 second clause"). For example, there may be a clause where the minimum time segment length is 2 seconds (referred to as a "2 second clause"). For example, there may be a clause where the minimum time segment length is 3 seconds (referred to as a "3 second clause"). A clause where the minimum time segment length is 6 seconds is referred to as a "6 second clause". Some examples of these clauses are listed in Table 1.

[0103] Table 1:

[0104]

[0105]

[0106] Some additional 3-second clauses are found in Table 2.

[0107] Table 2:

[0108]

[0109] Some additional 6-second clauses can be found in Table 3.

[0110] Table 3:

[0111]

[0112]

[0113] According to various implementations described herein, the defibrillator starts by analyzing the shortest interval clause available (e.g., a 1 second clause in the above example). However, the shortest interval can be as short as 100 milliseconds. After analyzing the ECG in the shortest segment to determine whether the ECG measurement or feature meets any of the shortest segment clause rule sets, if it is found that the criteria of no clause or rule set are met, the initial time period is extended, and the clause or rule set of the shorter interval is adjusted to provide a predetermined sufficient level of accuracy for the shock / no shock judgment for the extended time interval. Clause adjustments can take the form of relatively minor modifications to the threshold, or alternatively, new clauses or sub-clauses are added to maintain sufficient accuracy for different time intervals analyzed. Examples of these clauses can be found in Tables 1, 2, and 3.

[0114] In one example, the time extension is a 1 second interval (e.g., as described above with respect to the 1 second clause, the 2 second clause, the 3 second clause), or alternatively, as in the case of the 3 second clause and the 6 second clause, it may be 3 seconds. Alternatively, the time extension increment may be a subdivision of the minimum time increment (e.g., 0.1 second), thereby generating additional clauses of 1.1 second clauses, 1.2 second clauses, 1.3 second clauses, 1.4 second clauses, etc. In this manner, the defibrillator reanalyzes the ECG for each subincrement interval (e.g., 0.1 second) to make a new shock / no-shock decision every 0.1 second.

[0115] The presence of a shockable rhythm or a non-shockable rhythm can be determined 308 based on one or more clauses. In some examples, the presence of a shockable rhythm can be determined based on a single time segment, for example, if a single clause is met based on the time segment and the clause is associated with a shockable state, then the determination is made based only on the clause without further analysis that would delay treatment. In some examples, at least two or three time segments are used to make the determination, for example, unless at least one clause representing a shockable state is met for each of the two or three time segments, the clause continues to be evaluated for the data. If a particular clause is not met at a given time, a check 309 can be performed to see if other clauses defined about the calculated parameters are available to obtain more information to determine whether a shock decision or a no-shock decision should be made. If one or more other clauses that can provide this added information are available, processing includes accessing 306 clauses to repeat the check for the presence of a shockable rhythm or a non-shockable rhythm. On the other hand, if there are other clauses, then additional portions of the ECG data stored in the buffer may be processed 304. In some implementations, this may include repeating at least a portion of the above processing for additional ECG data that has been accumulated in the buffer since the last round of processing. In some implementations, this may include data that has been processed in the last round and new data that has been accumulated. In some implementations, this may include processing data corresponding to segments that do not overlap with segments processed in the last round.

[0116] In determining the presence of a shockable or non-shockable rhythm, the process 300 may include checking whether the accuracy of the clause satisfied by 310 exceeds a threshold. In some examples, the threshold is at least 99% accuracy. In some implementations, the check may also be performed before determining whether a shockable or non-shockable rhythm exists. If the accuracy of the clause exceeds the threshold (e.g., if the clause is a high accuracy or high confidence value), a feedback signal regarding the presence of a shockable or non-shockable rhythm may be presented to the caregiver (312), but such feedback signal is not required (e.g., when set to a specific setting, a defibrillation shock may be automatically given without user intervention). Although not required for some instances, the feedback may include generating a control signal that causes feedback to be presented on a device (e.g., AED 100, a mobile device associated with a caregiver, or a wearable device associated with a caregiver). The feedback may be presented, for example, in the form of a visual indication on a display screen, an audible sound presented via a speaker, or tactile feedback presented via a wearable or mobile device. Then, in response to the feedback, for example, if the feedback indicates that the patient is in a shockable state, the caregiver can choose to shock the patient. In some examples, once the capacitor of the defibrillator is charged, feedback can be provided so that the shock can be applied as quickly as possible. For example, the capacitor can be charged in approximately 5 seconds, where the 5 seconds can be enough time to determine the patient's state using the techniques described herein. Other slower techniques (e.g., techniques that only use voting schemes based on analyzing ECG segments of predetermined duration) may use more time to make such a determination, so that there is often a delay between the time the capacitor is charged and the time the determination of the patient's state is made.

[0117] If the accuracy of the clause does not exceed the threshold, the additional portion of the ECG data stored in the buffer can be processed 304. In some implementations, this can include repeating at least a portion of the above-mentioned processing for the additional ECG data accumulated in the buffer since the last round of processing. This operation can continue until a shockable or non-shockable rhythm is determined with the desired accuracy. By processing the data substantially continuously (e.g., once every several machine cycles, or several times per second), processing 300 allows rapid detection of shockable or non-shockable rhythms in the monitored ECG data. Since the processing does not rely on a fixed segment of data of a predetermined length, it is possible to quickly make a judgment without having to wait for the accumulation of data of a predetermined length. By using a clause with known specificity, the judgment process can be terminated as soon as a high confidence judgment is made, thereby saving precious time that can provide life-saving benefits to patients in critical conditions.

[0118] In some implementations, instead of using only a voting scheme (wherein a predetermined number of ECG segments are analyzed and the segments are marked as shockable or non-shockable), a technique that uses the techniques described herein to determine the presence of a shockable or non-shockable rhythm can be used. The following table shows the comparative performance of the two techniques based on a set of example experiments conducted using the same database of patient data. In the table below, the technique described herein is denoted as "Quick Analysis". In contrast, the voting scheme used in the comparison uses a maximum of three segments of ECG data, and is denoted as "3-out-2 voting" in the table below.

[0119] Table 4: Performance comparison

[0120] algorithm Sensitivity % Specificity% 3-choice-2 voting 99.67 99.86 Quick Analysis 99.67 99.87

[0121] Table 5: Comparison of analysis time

[0122] algorithm 3 seconds analysis 6-second analysis 9 Seconds Analysis 3-choice-2 voting 0％ 98％ 2％ Quick Analysis 85％ 14％ 1％

[0123] In some implementations, in a reconfirmation analysis mode (RAM), the determination of the presence of a shockable or non-shockable rhythm is performed during the administration of CPR. The following table shows the comparative performance of existing RAM techniques and the rapid analysis techniques described herein. In these experiments, the rapid analysis was again compared to the RAM analysis using a database containing ECG waveforms with CPR artifacts followed by a segment of a CPR free waveform.

[0124] Table 6: Performance comparison

[0125] algorithm Sensitivity % Specificity% Quick Analysis 97 99 RAM 96 99

[0126] Table 7: Analysis time comparison

[0127] algorithm 3 seconds analysis 6-second analysis 9 Seconds Analysis Quick Analysis 76％ 22％ 2％ RAM 62％ 36％ 2％

[0128] In some implementations, RAM analysis can be combined with the rapid analysis described herein. In such analysis, a high accuracy clause can be given a higher priority than the analysis during CPR. In a particular set of experiments, a three-second post-CPR segment is analyzed to calculate parameters. If the parameters meet the high confidence clause, the results from the analysis during the CPR result are ignored and the high accuracy clause is used for the final decision. The results of a particular set of experiments are provided in the following table.

[0129] Table 8: Performance

[0130] algorithm Sensitivity % Specificity% RAM+ Quick Analysis 96 99

[0131] Table 9: Analysis time

[0132] algorithm 3 seconds analysis 6-second analysis 9 Seconds Analysis RAM+ Quick Analysis 87％ 12％ 1％

[0133] Figure 4B A diagram is shown showing a technique 320 for determining whether a patient is in a shockable state or in a non-shockable state based on an ECG signal measured from the patient based on a voting process or based on whether a high accuracy clause is met before the voting process is completed. As described above, a high accuracy clause is a clause that exhibits a high predictive certainty based on PPV (positive predictive value) and / or NPV (negative predictive value). Technique 320 can be implemented as a defibrillator (e.g., Figure 1 The functions of the AED 100 shown.

[0134] The technique 320 includes two modes. The first mode 322 (sometimes referred to as a "continuous analysis mode" (CAM)) as an optional mode includes a CPR window 324, wherein in the CPR window 324, measurements are taken while CPR (chest compressions) are being applied to the victim. The second mode 326 (sometimes referred to as a "reconfirmation analysis mode" (RAM)) is performed during the no-CPR window 328, for example, during the no-CPR window 328, CPR (chest compressions) is not applied to the patient. In some examples, the application of CPR may interfere with the detection of a "clean" ECG signal (e.g., a signal without a lot of noise introduced by CPR therapy), so the second mode 326 can be used to detect a no-CPR signal. The technique 320 can alternate between the first mode 322 and the second mode 326, for example, in a rescue situation. In some examples, the mode changes when the AED 100 detects that CPR has stopped or started (e.g., by detecting that the movement associated with chest compressions has stopped or started). In some examples, the mode changes where the AED 100 provides instructions to the rescuer (eg, using an output device such as a display or an audio output device) to stop or start CPR.

[0135] In a first mode 322, continuous segments 330a-330c of the ECG signal during CPR are analyzed. For example, if any of these segments 330a-330c meet any clause used for ECG analysis during CPR, a decision (e.g., a CAM decision) is saved 332 as to whether the ECG signal represents a shockable rhythm or a non-shockable rhythm (e.g., for subsequent comparison with analysis during a no-CPR window to complete the determination). The decision can be based on the clauses applied to segments 330a-330c, for example. Because analysis during CPR may be less reliable than analysis performed when CPR is stopped, the saved decision can be confirmed in a second mode 326.

[0136] In the second mode 326, CPR is paused and ECG signal segments 334a-334c can be analyzed in the absence of any potential noise from CPR therapy. For example, the first segment 334a is analyzed by applying 336 a high accuracy clause to the segment 334a. If the high accuracy clause is met, for example, by indicating a shockable or non-shockable rhythm, a decision of a shockable or non-shockable rhythm is obtained 338, and no further analysis is required. If the high accuracy clause is not met, other normal accuracy clauses can be applied 340 to the segment 334a. If the segment satisfies one of the other clauses (e.g., a normal accuracy clause) and the clause indicates the same result (e.g., a RAM decision indicating a shockable rhythm or a non-shockable rhythm) as indicated in the saved decision 332 (e.g., a CAM decision) of the first mode 322, a decision of a shockable or non-shockable rhythm is obtained 342. In other words, since the results match, the matching result (e.g., between the CAM decision and the RAM decision) can be used as the final decision. However, if the other normal accuracy clauses indicate a different result than the saved decision, then the results of the other normal accuracy clauses are saved 344 and the technique 320 proceeds to the analysis of the second section 334b.

[0137] During the analysis of the second segment 334b, the high accuracy clause may be applied again 346. If the high accuracy clause is met, for example, by indicating a shockable or non-shockable rhythm, a decision of a shockable or non-shockable rhythm is made 348, and no further processing is performed. If the high accuracy clause is not met, other clauses may be applied 350 to segment 334b. If another clause (e.g., a normal accuracy clause) indicates the same result as the saved result 344 of the analysis of the first segment 334a (e.g., a shockable or non-shockable rhythm), a decision of a shockable or non-shockable rhythm is made 352. If the other normal accuracy clause indicates a different result than the saved result 344, the result of the other normal accuracy clause is saved 354, and the technique 320 proceeds to the analysis of the third segment 334c.

[0138] During the analysis of the third segment 334c, the high accuracy clause is again applied 356. If the high accuracy clause is met, for example, by indicating a shockable or non-shockable rhythm, a decision of a shockable or non-shockable rhythm is obtained 358. If the high accuracy clause is not met, other normal accuracy clauses may be applied 360 to segment 334c, and a "2 out of 3" vote may occur. If another clause (e.g., a normal accuracy clause) indicates the same result (e.g., a shockable rhythm or a non-shockable rhythm) as the saved result 344 of the analysis of the first segment 334a or the saved result 354 of the analysis of the second segment 334b, a decision of a shockable or non-shockable rhythm is obtained 364. In other words, in the event that two of the three results match, these matching results (e.g., shockable or non-shockable) are used as a decision as to whether the patient is in a shockable state or in a non-shockable state.

[0139] Figure 4C A diagram is shown showing a technique 370 for determining whether a patient is in a shockable state or a non-shockable state based on an ECG signal measured from the patient according to a voting process or according to whether a high accuracy clause is met before the voting process is completed. Technique 370 can be implemented as a defibrillator (e.g., Figure 1 100).

[0140] Technique 370 includes a single mode 372 in which measurements are taken while CPR is being applied to the victim during a CPR window 374. In this mode 372, ECG signal segments 376a-376c may be analyzed. For example, a first segment 376a is analyzed by applying 378 a high accuracy clause to the segment 376a. If the high accuracy clause is met, for example by indicating a shockable or non-shockable rhythm, a shockable or non-shockable rhythm determination is made 380. If the high accuracy clause is not met, other normal accuracy clauses may be applied to segment 334a. If the normal accuracy clause indicates that the rhythm is shockable or non-shockable, the result is saved 382 and technique 370 proceeds to analysis of a second segment 376b.

[0141] During the analysis of the second segment 376b, the high accuracy clause may be applied again 384. If the high accuracy clause is met, for example, by indicating a shockable or non-shockable rhythm, a shockable or non-shockable rhythm determination is obtained 386. If the high accuracy clause is not met, other clauses may be applied 388 to segment 376b. If the clause (e.g., a non-high accuracy clause) indicates the same result (e.g., shockable rhythm or non-shockable rhythm) as the saved result 382 of the analysis of the first segment 376a, a shockable or non-shockable rhythm determination is obtained 390. If the analysis using other clauses indicates a different result than the saved result 382, ​​the analysis results of the other clauses are saved 392 and the technique 370 proceeds to the analysis of the third segment 376c.

[0142] During the analysis of the third segment 376c, the high accuracy clause is again applied 394. If the high accuracy clause is met, for example, by indicating a shockable or non-shockable rhythm, a decision of a shockable or non-shockable rhythm is obtained 396. If the high accuracy clause is not met, other clauses (e.g., normal accuracy clauses) can be applied 398 to segment 396c, and a "3 out of 2" vote can occur. If the other clauses (e.g., normal accuracy clauses) indicate the same result (e.g., shockable rhythm or non-shockable rhythm) as either the saved results 382 of the analysis of the first segment 376a or the saved results 392 of the analysis of the second segment 376b, a decision of a shockable or non-shockable rhythm is obtained 399. In other words, in the event that two of these results match, these matching results (e.g., shockable or non-shockable) are used as a decision as to whether the patient is in a shockable state or in a non-shockable state based on the measured ECG signal.

[0143] Figure 4D A flow chart showing an example process 400 for identifying the presence of a shockable rhythm in ECG data. The process may be performed, for example, by Figure 2 The logic module 218 shown is implemented and applied to utilize Figure 3 The ECG waveform 250 represents data such as ECG data.

[0144] Processing 400 includes receiving 402 data representing an ECG signal of a patient, for example, while the patient is undergoing rescue therapy. Processing 400 includes processing 404 variable length time segments of the ECG signal. The variable length time segments include an initial time segment (e.g., between 0.1 seconds and 3.0 seconds) that is incrementally extendable using one or more additional time segments (e.g., 0.1 seconds, 0.5 seconds, 1.0 seconds, 1.5 seconds, 2.0 seconds, 2.5 seconds, 3.0 seconds, etc.). In this way, the amount of time required to determine whether the patient is in a shockable state or a non-shockable state is unknown prior to the determination. In some examples, the length of the time segment being processed is changed based on the identification of the feature being used to determine whether the ECG rhythm is shockable or non-shockable.

[0145] The processing includes identifying 406 one or more features of the variable length time segment based on an initial rule set corresponding to the initial time segment. In some examples, the initial rule set includes one or more clauses representing constraints on one or more features of the ECG signal, wherein each of the one or more features of the ECG signal represents a characteristic of a waveform of the ECG signal within the processed time segment.

[0146] The processing includes determining 408 whether an ECG rhythm represented by an ECG signal is shockable or non-shockable based on an initial rule set, or determining whether to extend a variable length time segment for further analysis (e.g., if an accurate determination cannot be made using the initial time period). The processing includes extending 410 an initial time segment of the variable length time segment using one or more additional time periods (e.g., 0.1 seconds, 0.5 seconds, 1.0 seconds, 1.5 seconds, 2.0 seconds, 2.5 seconds, 3.0 seconds, etc.). The processing includes identifying 412 one or more features of the variable length time segment based on an adjusted rule set corresponding to the adjusted time period to update the determination of whether the ECG rhythm is shockable or non-shockable. In some examples, the initial rule set may include at least one high accuracy clause, and the adjusted rule set includes at least one lower accuracy clause, such that the at least one high accuracy clause is different from the at least one lower accuracy clause.

[0147] Process 400 includes providing 414 an output based on the determination, such as by using an output circuit or output module (eg, an output screen, an audio circuit, etc.).

[0148] In some examples, data representing the patient's ECG signal is received while CPR is being performed. In some examples, data representing the patient's ECG signal is received when CPR is temporarily stopped. For example, processing 400 may include identifying whether an interruption of chest compressions occurs, and analyzing variable length time segments of the ECG signal according to an initial rule set during the interruption of chest compressions. In some examples, identifying whether an interruption of chest compressions occurs includes processing a signal generated from at least one of an ECG electrode and a motion sensor.

[0149] Figure 4E A flow chart showing an example process 420 for identifying the presence of a shockable rhythm in ECG data. The process 420 may be performed, for example, by Figure 2 The logic module 218 shown is implemented and applied to utilize Figure 3 ECG data such as the data represented by the ECG waveform 250.

[0150] Processing 420 includes receiving 422 data representing an ECG signal of a patient while the patient is undergoing rescue treatment. Processing 420 includes processing 424 a variable length time segment of an ECG signal including an initial time segment between 0.1 seconds and approximately 3 seconds. The processing includes, for example, identifying 426 whether an interruption of chest compressions occurs by processing a signal generated from at least one of an ECG electrode and a motion sensor. The processing includes analyzing 428 a variable length time segment of an ECG signal during an interruption of chest compressions according to an initial rule set based on the initial time segment. The analysis can be used to determine whether the ECG rhythm represented by the ECG signal is shockable or non-shockable. In some examples, the initial rule set includes one or more clauses representing constraints for one or more features of the ECG signal, wherein each of the one or more features of the ECG signal represents a characteristic of a waveform of the ECG signal within the processed time segment.

[0151] The processing includes determining 430 whether the ECG rhythm represented by the ECG signal is shockable or non-shockable based on the initial rule set, or (e.g., if an accurate determination cannot be made using the initial time period) determining whether to extend the variable length time segment for further analysis. The processing includes generating 432 an output based on the determination whether the ECG rhythm represented by the ECG signal is shockable or non-shockable. The processing 420 includes providing 434 an output based on the determination, for example, by using an output circuit or output module (e.g., an output screen, an audio circuit, etc.).

[0152] Figure 4F Shown with the ability to use, for example Figure 4B to Figure 4CAn example of a waveform 436 of two segments 438a-438b analyzed by one of the techniques 320, 370 is shown. For example, the first segment 438a does not clearly meet any high accuracy clauses, so the analysis using the techniques 320, 370 will proceed to the analysis of the second segment 438b, for example. In the example shown, the second segment 438b clearly meets the high accuracy clauses, so a decision can be made based on the high accuracy clauses indicating whether the waveform 436 is shockable or non-shockable. In this case, the second segment 438b provides a stronger indication that the ECG rhythm is non-shockable.

[0153] As discussed herein, in order to judge as quickly as possible whether the ECG rhythm represented by the ECG signal is shockable or non-shockable, it may be preferred to analyze the gap or interruption of chest compression as quickly and efficiently as possible. For example, at any point whenever chest compression stops, even for a short period of time, it may be appropriate to judge whether the rhythm is shockable or non-shockable or to judge whether more time is needed. In some instances, the interruption of chest compression may be relatively short (e.g., less than approximately 3 seconds). For example, a rescuer may want to readjust his / her position, exchange roles with another rescuer, be distracted, be slightly tired, or may suspend chest compression. For these situations, it may be advantageous for the system to start its rhythm analysis immediately because it is uncertain when chest compression will restart, so that the time spent for determining the patient's heart state is the shortest. In some instances, the time period of chest compression interruption may be quite large (e.g., greater than 12 to 15 seconds). For example, the rescuer may be instructed to stop CPR chest compressions and wait for the defibrillator system to perform ECG analysis, and the rescuer may completely switch to another resuscitation activity such as ventilation, etc. In these cases, the ECG analysis may run its normal course.

[0154] For various implementations, as discussed above, the defibrillator system can be configured to start ECG analysis once it identifies the interruption of chest compressions being applied to the patient or shortly after identification. The defibrillator system can track chest compressions using any appropriate method or technology. In some implementations, the defibrillator system includes a motion sensor (e.g., accelerometer, velocity sensor, displacement sensor) at the position where the rescuer is applying chest compressions to detect the presence of chest compressions. For example, an accelerometer can be embedded in a chest compression sensor, and the rescuer can place the chest compression sensor between the patient's chest and his / her hand during CPR compressions. The acceleration signal generated by the sensor can be processed accordingly. This motion sensor can also be used to sense the depth and rate of chest compressions to provide appropriate CPR feedback to the rescuer. Various implementations of determining whether a cardiac rhythm is shockable based solely on time periods of the ECG signal during which no CPR chest compressions were delivered are described in U.S. Patent 6,961,612, entitled “CPR Sensitive ECG Analysis in an Automatic External Defibrillator,” and may be used in conjunction with the systems and methods described herein; the entirety of which is incorporated herein by reference.

[0155] Figure 4G to Figure 4LThe description includes an illustrative implementation of the following schematic diagram, in which the defibrillator system quickly identifies whether an interruption of chest compression occurs and applies an appropriate analysis algorithm. In each implementation, each schematic diagram shows a period of time when CPR chest compressions are being provided. In this CPR window, as further described herein, the defibrillator system can optionally apply a continuous analysis consultation for whether a shockable or non-shockable rhythm is detected while considering the artifacts produced by chest compressions. This continuous analysis consultation can include appropriate filtering, frequency-based analysis, and / or other appropriate analysis techniques for providing an indication of whether the rhythm represented by the ECG signal is shockable or non-shockable. The indication can be used to determine (subsequently or during analysis) whether a defibrillation shock should be applied. For example, a continuous analysis consultation can indicate that the ECG rhythm is likely to be shockable, and this indication can then be confirmed via a subsequent "hands-free" ECG analysis (i.e., no CPR chest compressions occur). Typically, a "hands-free" ECG analysis can provide a more accurate shock analysis compared to a continuous ECG analysis during compressions. Examples of suitable continuous analysis advisory algorithms (during compressions) that may be employed include those described in U.S. Patent 8,706,214, entitled "ECG Rhythm Advisory Method," and U.S. Patent 8,880,166, entitled "Defibrillator Display," each of which is incorporated herein by reference in its entirety.

[0156] As discussed above, in order to make a final judgment about whether the ECG rhythm sensed from the patient is an ECG rhythm to which a defibrillation shock should be applied, it may be necessary to stop chest compressions for a short period of time to analyze a cleaner ECG (e.g., without artifacts caused by CPR chest compressions). Therefore, the defibrillation system prompts the user to stop CPR, for example, via an audio and / or visual prompt from the user interface of the defibrillator. If the user confirms the prompt and interrupts the process of applying CPR chest compressions, the system will then analyze the clean ECG (without chest compressions) to determine whether there is a shockable or non-shockable rhythm. However, the user may not confirm the prompt to stop CPR (e.g., may not see / hear, or may ignore the prompt from the defibrillator) and continue chest compressions. If the user continues chest compressions, the system may continue to apply a continuous analysis consultation taking into account chest compression artifacts in the ECG despite the prompt to stop chest compressions. However, once the user pauses chest compressions, the system can immediately or within a short period of time switch the type of ECG analysis from a continuous analysis consultation (with compressions) to a "hands-off" analysis mode (without compressions) that can more accurately confirm whether a shockable or non-shockable rhythm is present.

[0157] like Figure 4GAs shown, in the event that the defibrillator system determines that chest compressions should be stopped to analyze the ECG signal 440 to determine whether a shock should be applied, an instructional prompt 442 is issued to the rescuer to stop CPR and remove hands from the patient. In various implementations, the system may optionally pause for a short period of time before "stopping" the ECG analysis. This short pause 444 may be preferred in some cases to ensure that the ECG signal is substantially free of artifacts, presence or residue caused by chest compressions. Although Figure 4G The example shows a short pause time of approximately 1.5 seconds before a "hands-off" ECG analysis, but any suitable pause time may be used, such as less than 2 seconds, less than 1.5 seconds, less than 1 second, less than 0.5 seconds, less than 0.2 seconds, less than 0.1 seconds, etc. In some implementations, although not shown in the figures, a "hands-off" ECG analysis advisory may be immediately employed when an interruption in chest compressions is detected.

[0158] Figure 4H An implementation is described in which chest compressions 446 are stopped even before the system provides an instructional prompt. Although it is not advisable to stop chest compressions prematurely, rescuers often do so for a variety of reasons. In this case, although not shown in the figure, the system can prompt the rescuer to continue chest compressions or display an idle timer showing how long the rescuer has stopped compressions until the system determines that chest compressions should be interrupted to begin a "hands off" analysis. Figure 4H The implementation still pauses for a short period of time after the prompt to stop CPR, however, it is understood that such a pause is not required. For example, upon detecting an interruption in chest compressions, the system can automatically and / or immediately start a "hands-off" ECG analysis consultation without such a pause.

[0159] Fig. 4I An implementation is shown in which chest compressions continue for a short time 448 (e.g., 1-2 seconds) after the system issues an instructional prompt to stop CPR. In this case, the system tracks chest compressions until the compressions stop, optionally pauses for a short period of time (e.g., approximately 1.5 seconds), and then begins a "hands-off" ECG analysis advisory according to the present disclosure.

[0160] Figure 4J Another implementation of the following example is described, in which Fig. 4ICompared with the situation shown in the figure, chest compressions are continued for a longer period of time 450. Here, the system continues to track chest compressions, and then after a sufficiently long period of time, the system issues a subsequent instructive prompt to remind the rescuer to stop CPR. It should be understood that the subsequent instructive prompt can be provided at any appropriate time, wherein the appropriate time can be predetermined by an appropriate time interval. In this particular case, the system issues a subsequent prompt after approximately 3 seconds (wherein in some cases, the approximately 3 seconds can be the same as the initial time period of the ECG analysis). As further shown, in the case of interruptions in chest compressions, the system optionally pauses for a short time (e.g., approximately 1.5 seconds) and then starts a "stop" ECG analysis consultation.

[0161] Figure 4K An implementation is shown in which chest compressions continue for an even longer period of time 452. In this example, the system issues a first prompt 454 instructing the user to stop chest compressions, and then after a sufficiently long time interval 456 (e.g., approximately 3 seconds), the system issues a subsequent instructional prompt 458 instructing the rescuer to stop CPR compressions. Here, however, the system continues to sense chest compressions after the subsequent instructional prompt, resulting in a further delay in the “hands off” CPR analysis advisory. In various implementations, the ECG signal is tracked according to short time segments (e.g., approximately 3 seconds), and once the system identifies an interruption in chest compressions, the “hands off” CPR analysis advisory begins at the beginning of the next time segment. As shown in Figure 4K As more specifically shown in , the system does not detect an interruption in chest compressions until approximately 2 seconds after the most recent instructional prompt to stop CPR compressions. Then, after approximately 1 second of the remaining 3-second interval has passed, the "hands off" CPR analysis advisory begins. However, for certain implementations, after a subsequent instructional prompt, once an interruption in chest compressions is determined, the system can immediately begin the "hands off" CPR analysis advisory without the need for an optional pause.

[0162] Figure 4L Another exemplary implementation is shown where no interruption of CPR chest compressions is detected despite the presence of multiple instructional prompts 460, 462 to stop chest compressions. In this case, a "hands-off" CPR analysis advisory cannot be used because chest compressions remain uninterrupted. In this case, a continuous analysis advisory algorithm (taking into account chest compression artifacts) is applied throughout the time chest compressions are being administered.

[0163] Some examples of parameters that appear in clauses are provided in Table 10.

[0164] Table 10:

[0165] parameter describe Widthvar Variability of the width of the QRS complex Width Width of the QRS complex qrsv Variability of RR intervals maxamp The maximum amplitude of the ECG waveform minamp The absolute minimum amplitude of the ECG waveform qrsr Average RR interval svtbeats The number of SVT beats detected in the waveform

[0166] Figures 5A-5BExamples of waveforms 500a-500b having characteristics corresponding to some of the parameters listed in Table 10 are shown.

[0167] Figure 5A An example waveform 500a is shown demonstrating how the values ​​of qrsr, widthvar, and qrsv can be calculated. For example, the waveform 500a includes three QRS complexes 502a-502c having widths (measured in millimeters) identified as width1, width2, and width3. In addition, the distances (measured from R to R) between the QRS complexes 502a-502c are identified by qrsr1 and qrsr2. The parameter qrsr can be calculated by averaging the measured values ​​of the qrsr values ​​(e.g., qrsr1 and qrsr2). The parameter widthvar can be calculated by calculating the variability of the measured values ​​of the width (e.g., width1, width2, and width3). The parameter qrsv can be calculated by calculating the variability of the qrsr values ​​(e.g., qrsr1 and qrsr2). The variability can be determined by sorting the relevant values ​​(e.g., the values ​​of width and the values ​​of qrsr) in order of size and grouping them by intervals (e.g., 20 milliseconds). The number of intervals thus obtained corresponds to the variability.

[0168] Figure 5B An example waveform 500b is shown that demonstrates how to calculate the value of svtbeats. The waveform 500b has eight supraventricular tachycardia-type heartbeats 510a-510h.

[0169] One or more parameters calculated based on the ECG data (e.g., the parameters provided in Table 10) may be used to define various high accuracy clauses and normal accuracy clauses used in identifying shockable and non-shockable rhythms in a segment of ECG data. Fig. 6A An example waveform 600a of a shockable waveform for ventricular tachycardia is shown in . The 3 second clause is satisfied (met) when the QRS width is greater than 120ms and the degree of width variability that may represent a wide complex rhythm is limited. The clause represents a series of constraints on parameters corresponding to characteristics of the ECG signal. In some cases, if the rhythm is VT, a wide complex rhythm with an RR interval of less than 400ms may be shockable. However, for some cases, additional checks may be required to determine the presence of VT with high accuracy. A check of whether Svtbeats is less than 3 counts excludes the rhythm from representing SVT. Therefore, once all conditions of the clause are met, the presence of a shockable VT can be identified with high accuracy.

[0170] exist Figure 6BAn example non-shockable waveform 600b that complies with the 3 second clause is shown in FIG. 600b. This clause is satisfied in cases where the average width of the QRS complex is small and the width variability is limited, conditions that may indicate a narrow complex rhythm. Since the average amplitude is greater than 300 μV, the chance of a QRS detection error is low. Thus, in cases where the complex is narrow and the RR interval is >265 ms, the rhythm corresponding to the underlying ECG data can be identified as non-shockable with high specific accuracy.

[0171] In another implementation of a clause for determining whether an ECG signal is characteristic of a normal sinus rhythm (which returns a non-shockable result), the clause evaluates whether the ECG signal has a single distinct peak representing a normal QRS. The clause can identify whether the maximum slope of the signal is greater than a certain threshold and / or whether the signal exhibits a relatively flat time period (wherein the slope of the analyzed time segment remains below another threshold). For example, if the maximum slope of the time segment appears to be greater than 50-100 μV (e.g., or greater than 100-150 μV, greater than 150-200 μV, etc.), and / or if the slope of the analyzed time segment remains below 30 μV / ms (e.g., below 20 microvolts / milliseconds, below 10 μV / ms, etc.) within a time period greater than 50 ms (e.g., or greater than 100 ms, greater than 150 ms, greater than 200 ms, etc.), then the peak characteristic of a normal sinus rhythm can be positively identified. Examples of such clauses may be evaluated for any suitable time period (eg, 0.5 seconds, 1 second, 1.5 seconds, 2 seconds, 2.5 seconds, 3 seconds, etc.).

[0172] In another implementation of a clause for determining whether an ECG signal is characteristic of an asystolic rhythm (which returns a non-shockable result), the ECG signal is evaluated for substantially low amplitudes. Here, the clause can identify whether the magnitude of the maximum or minimum amplitude of the signal is less than a specific threshold. For example, if the magnitude of the maximum amplitude is less than 150 μV (e.g., or less than 100 μV, less than 50 μV, etc.) and / or the magnitude of the minimum amplitude is less than 150 μV (e.g., or less than 100 μV, less than 50 μV, etc.), it can be determined to be a flatline rhythm. Examples of such clauses can be evaluated for any appropriate time segment (e.g., 0.5 seconds, 1 second, 1.5 seconds, 2 seconds, 2.5 seconds, 3 seconds, etc.).

[0173] In the implementation of a clause for determining whether an ECG signal is a characteristic of a VF type rhythm (which returns a shockable result), the ECG signal is evaluated for multiple peaks that last for a sufficiently long period of time. The clause can be configured to identify whether the maximum slope of the ECG signal is greater than a specific threshold, whether the signal is almost flat, and / or whether the waveform crosses the zero point multiple times. For example, the clause may seek to identify whether the maximum slope of the time segment is greater than 20μV (e.g., or greater than 50μV, greater than 100μV, etc.). The clause can also analyze the signal to determine whether the slope of the analyzed time segment remains below 30μV / millisecond (e.g., or less than 20 microvolts / millisecond, less than μV / ms, etc.) within a time period of less than 150ms (e.g., or less than 100ms, less than 50ms, etc.). In addition, the clause can analyze the signal to determine whether the waveform crosses the zero point more than 5 times (e.g., or greater than 10 times, greater than 20 times, greater than 30 times, etc.). Examples of such clauses may be evaluated for appropriate time periods (eg, 1 second, 1.5 seconds, 2 seconds, 2.5 seconds, 3 seconds, etc.).

[0174] In the implementation of another clause that evaluates whether the ECG signal is a VT-type rhythm, the clause determines whether the ECG signal exhibits a high frequency with a wide complex wave. In particular, the clause can evaluate the signal to identify whether the average peak-to-peak interval of the waveform is less than a specific threshold, whether the average width of the QRS complex wave is greater than a threshold, and / or whether the overall width variation of the QRS complex wave is relatively regular. For example, the clause can evaluate whether the average peak-to-peak interval of the ECG signal within the analyzed time segment is less than 500ms (e.g., less than 400ms, less than 350ms, less than 300ms, less than 250ms, etc.). The clause can also determine whether the average width of the QRS complex wave is greater than 50ms (e.g., or greater than 100ms, greater than 150ms, greater than 200ms, etc.), and evaluate whether the variability of the width of the QRS complex wave is substantially regular.

[0175] In another implementation of a clause for determining whether an ECG signal is characteristic of a VT rhythm (which returns a shockable result), the ECG signal is evaluated for a fast heart rate with a wide QRS complex over a period of 4 to 8 seconds (e.g., approximately 6 seconds). The clause can be configured to identify whether the heart rate is greater than a certain threshold (e.g., greater than 120 bpm, greater than 130 bpm, greater than 140 bpm, greater than 150 bpm, greater than 160 bpm, etc.), whether the interval between peaks is less than a threshold (e.g., less than 600 ms, less than 500 ms, less than 400 ms, less than 300 ms, etc.), whether the width of the QRS complex is greater than a threshold (e.g., greater than 120 ms, greater than 130 ms, greater than 140 ms, greater than 150 ms, greater than 160 ms, etc.), whether the overall width variation of the QRS complex is relatively regular, and / or whether the signal exhibits little flatness (e.g., the ECG remains below 50 μV / ms over the analyzed time segment over a period of greater than 100 ms). Furthermore, as discussed herein, the systems and methods of rapid ECG analysis described herein may involve incrementally extending the time period for ECG analysis in the event of uncertainty as to whether the ECG rhythm is shockable or non-shockable. Figure 6C , the ECG analysis may begin at the first region 602 of the ECG signal 600c, which is shown to last approximately 1 second. However, in this example, the ECG signal may not provide enough information for the clause associated with the 1 second time segment to make a final decision of shockable or non-shockable. Therefore, the time segment of the ECG analysis may be extended in increments to collect more information to make a final decision. Figure 6C As shown in the second region 604 of FIG. 6 , the variable length time segment is extended by another second. In the extended analysis window, the QRS waveform characteristics of normal sinus are identified, so that a non-shockable ECG rhythm is identified.

[0176] As described above, Table 2 includes a number of exemplary clauses that can be used as 3 second clauses for an initial rule set. As an example, for an initial time period (e.g., 3 seconds) of ECG analysis, if any of the clauses marked as high accuracy are met, a decision is made as to whether the rhythm is shockable or non-shockable without further analysis. As discussed herein, it should be understood that the initial time period can span any appropriate time period. For example, the same exemplary clauses as those provided in Tables 1 and 3 for other time segments of analysis can be applied.

[0177] Table 3 includes multiple exemplary clauses that can be used as 6-second clauses for the adjusted rule subset. Continuing with the above example, if no high accuracy clause is met, the time period of analysis continues further and the adjusted rule set is applied. Thus, the initial time period can be extended to the adjusted time period (e.g., 6 seconds) for ECG analysis. As discussed for a specific implementation, for this adjusted time period, the extended duration segment can be further subdivided and the new clause or rule set can adopt these techniques known to those skilled in the art, such as voting models. In this case, the new adjusted longer duration clause can be applied separately to multiple time segments within the span of the adjusted time period (e.g., multiple 3-second time segments, totaling 6 to 9 seconds or 6 to 12 seconds of ECG analysis). For each time segment, a vote for shockable or non-shockable is returned, and the decision of shock or non-shock is based on the result of the vote.

[0178] In various implementations, the analysis algorithm may employ inputs from sensors other than sensors for sensing the electrical activity of the heart to provide information that will make a determination whether the patient is in a shockable state or a non-shockable state more quickly than without the additional information. For example, it is shown that a motion sensor (e.g., an accelerometer) can be used to measure heart wall motion and can also provide an indication of whether the patient's heart has achieved a normal sinus rhythm. This concept is described in International Patent Application PCT / US16 / 37081, entitled "Detection of Myocardial Contractions Indicative of Perfusion," filed on June 10, 2016, and is incorporated herein by reference in its entirety. If the additional sensor input provides evidence that the heart is beating mechanically regularly, the analysis algorithm can more quickly conclude that a defibrillation shock should not be applied to the patient. Alternatively, if the additional sensor shows that the heart is beating irregularly, the analysis algorithm may require more information to make a final determination as to whether the patient should be defibrillated.

[0179] It should be understood that the teachings of the present disclosure are generally applicable to external medical monitoring and / or treatment devices (e.g., devices that are not fully implanted in the patient's body). That is, the algorithms described herein for analyzing a patient's ECG to quickly determine whether it is appropriate to apply a defibrillation shock to the heart can be applied to external medical monitoring and / or treatment devices. External medical devices may, for example, include mobile medical devices that are capable of and designed to move with the patient as the patient goes about his or her daily life. Examples of mobile medical devices are wearable medical devices such as wearable cardioverter-defibrillators (WCDs), wearable cardiac monitoring devices, in-hospital devices such as in-hospital wearable defibrillators, short-term wearable cardiac monitoring and / or treatment devices, mobile telemetry devices, and other similar wearable medical devices.

[0180] The wearable medical device can be used continuously (e.g., substantially or almost continuously) by the patient. In some implementations, the continuous use can be substantially or almost continuous in nature. That is, the wearable medical device can be used continuously except for sporadic time periods when use is temporarily stopped (e.g., during the patient's bathing, during the patient's change of new and / or different clothing, during the charging / replacing of the battery, during the washing of clothing). Nevertheless, such substantially or almost continuous use is still considered to be continuous use. For example, the wearable medical device can be configured to be worn by the patient for up to 24 hours a day. In some implementations, the patient can remove the wearable medical device within a short portion of the day (e.g., within half an hour of bathing).

[0181] In addition, the wearable medical device may be configured as a medical device for long-term or extended use. These devices may be configured to be used by patients for extended periods of days, weeks, months, or even years. In some examples, the wearable medical device may be used by a patient for an extended period of at least one week. In some examples, the wearable medical device may be used by a patient for an extended period of at least 30 days. In some examples, the wearable medical device may be used by a patient for an extended period of at least one month. In some examples, the wearable medical device may be used by a patient for an extended period of at least two months. In some examples, the wearable medical device may be used by a patient for an extended period of at least three months. In some examples, the wearable medical device may be used by a patient for an extended period of at least six months. In some examples, the wearable medical device may be used by a patient for an extended period of at least one year. In some implementations, extended use may not be interrupted until a physician or other caregiver provides specific instructions to the patient to stop using the wearable medical device.

[0182] Regardless of the extended period of wear, use of the wearable medical device may include continuous or nearly continuous wearing by the patient as described above. For example, continuous use may include continuously wearing or attaching the wearable device to the patient, for example via one or more of the electrodes described herein, during the following two time periods: a period of monitoring, and a period of time during which the device may not be monitoring the patient but is still worn by the patient or otherwise attached to the patient. The wearable medical device may be configured to continuously monitor the patient's cardiac-related information (e.g., ECG information, including arrhythmia information, etc.) and / or non-cardiac information (e.g., blood oxygen, the patient's temperature, blood glucose levels, tissue fluid levels, and / or lung sounds). The wearable medical device may perform monitoring at periodic or non-periodic time intervals or times. For example, monitoring within an interval or time may be triggered by a user action or another event.

[0183] As described above, the wearable medical device can be configured to monitor other physiological parameters of the patient in addition to monitoring heart-related parameters. For example, the wearable medical device can be configured to monitor, for example, lung sounds (e.g., using a microphone and / or accelerometer), breathing sounds, sleep-related parameters (e.g., snoring, sleep apnea), tissue fluid (e.g., using a radio frequency transmitter and sensor), etc.

[0184] Other example wearable medical devices include automatic heart monitors and / or defibrillators for use in specific dedicated conditions and / or environments, such as in a war zone or within an emergency vehicle. These devices can be configured so that these medical devices can be used immediately (or substantially immediately) in a life-saving emergency. In some examples, the wearable medical devices described herein can have a pacing function, such as being able to provide therapeutic pacing pulses to a patient.

[0185] In implementation, the example therapeutic medical device may include an in-hospital continuous monitoring defibrillator and / or pacing device, such as an in-hospital wearable defibrillator. In this example, the electrode may be attached to the patient's skin in an adhesive manner. For example, the electrode may include a disposable adhesive electrode. For example, the electrode may include a sensing and treatment component disposed on a separate sensing and treatment electrode adhesive patch. In some implementations, both the sensing and treatment components may be integrated and disposed on the same electrode adhesive patch attached to the patient. In an example implementation, the electrode may include a front adhesively attachable treatment electrode, a back adhesively attachable treatment electrode, and a plurality of adhesively attachable sensing electrodes. For example, the front adhesively attachable treatment electrode is attached to the front of the patient's torso to deliver pacing or defibrillation treatment. Similarly, the back adhesively attachable treatment electrode is attached to the back of the patient's torso. In an example solution, at least three ECG adhesively attachable sensing electrodes may be attached to at least the upper chest of the patient near the right arm, the upper chest of the patient near the left arm, and toward the lower part of the patient's chest in a manner prescribed by a trained professional.

[0186] Patients being monitored with in-hospital defibrillators and / or pacing devices may be confined to a hospital bed or room for a significant amount of time (e.g., more than 90% of the time the patient stays in the hospital). As a result, the user interface may be configured to interact with a user other than the patient (e.g., a nurse) to perform device-related functions, such as initial device baseline alignment, setting and adjusting patient parameters, and replacing device batteries.

[0187] In an implementation, examples of therapeutic medical devices may include short-term continuous monitoring defibrillators and / or pacing devices, such as short-term outpatient wearable defibrillators. For example, such short-term outpatient wearable defibrillators may be prescribed by a physician for patients suffering from syncope. The wearable defibrillator may be configured to monitor patients suffering from syncope by, for example, analyzing the patient's cardiac activity for abnormal patterns that may indicate abnormal physiological function. For example, such abnormal patterns may occur before the onset of symptoms, during the onset of symptoms, or after the onset of symptoms. In this example implementation of a short-term wearable defibrillator, the electrode assembly may be adhesively attached to the patient's skin and have the same structure as the in-hospital defibrillator described above.

[0188] In some implementations, the medical device may be a patient monitoring device that does not have a treatment or therapy function. Such a patient monitoring device may be equipped with an algorithm described herein to determine whether the patient needs electrical treatment such as defibrillation. For example, such a patient monitoring device may include a heart monitoring device or a heart monitor, wherein the heart monitoring device or the heart monitor is configured to monitor one or more cardiac physiological parameters of the patient, such as to remotely monitor and / or diagnose the patient's condition. For example, these cardiac physiological parameters may include the patient's electrocardiogram (ECG) information and other related cardiac information. The heart monitoring device is a portable device that the patient can carry with him or her as he or she conducts their daily life. The heart monitor may be configured to detect the patient's ECG through a plurality of heart sensing electrodes. For example, the heart monitor may be attached to the patient via at least three adhesive heart sensing electrodes disposed on the patient's torso. These heart monitors are used in mobile cardiac telemetry (MCT) applications and / or continuous cardiac event monitoring applications, such as in a patient population reporting irregular cardiac symptoms and / or conditions. Example cardiac conditions may include atrial fibrillation, bradycardia, tachycardia, atrioventricular block, Lown-Ganong-Levine syndrome, atrial flutter, sinus node dysfunction, cerebral ischemia, syncope, atrial pauses and / or palpitations. For example, these patients may be prescribed to use a cardiac monitor for an extended period of time (e.g., 10 to 30 days or longer). In some mobile cardiac telemetry applications, a portable cardiac monitor may be configured to monitor a patient's cardiac abnormalities substantially continuously, and in the event of such an abnormality being detected, the monitor may automatically send data related to the abnormality to a remote server. The remote server may be located in a 24-hour manned monitoring center, where the data is interpreted by qualified cardiac trained reviewers and / or caregivers, and feedback is provided to the patient and / or designated caregiver via detailed periodic or event-triggered reports. In a specific cardiac event monitoring application, the cardiac monitor is configured to allow the patient to manually press a button on the cardiac monitor to report symptoms. For example, a patient may report symptoms such as: missed beats, shortness of breath, mild headache, fast heartbeat, fatigue, fainting, chest discomfort, weakness, dizziness, and / or vertigo. A cardiac monitor may record predetermined physiological parameters (e.g., ECG signals) of a patient for a predetermined amount of time (e.g., 1-30 minutes before and 1-30 minutes after the reported symptoms). A cardiac monitor may be configured to monitor physiological parameters of a patient other than cardiac-related parameters.

[0189] Example Wearable Medical Device

[0190] Figure 7An example medical device 700 is shown, where the medical device 700 is external, mobile, and wearable by a patient 702, and is configured to implement one or more of the configurations described herein. For example, the medical device 700 can be a non-invasive medical device configured to be substantially external to the patient. Such a medical device can, for example, be a mobile medical device that is capable of and designed to move with the patient as the patient goes about his or her daily life. For example, the medical device 700 as described herein can be attached to the patient's body, such as from Purchased by medical company Wearable cardioverter defibrillators, etc. Such wearable defibrillators are usually worn almost continuously or substantially continuously for 2 to 3 months at a time. While being worn by the patient, the wearable defibrillator can be configured to continuously or substantially continuously monitor the patient's vital signs, and when it is determined that treatment is required, it can be configured to deliver one or more therapeutic electrical pulses to the patient. For example, such therapeutic shocks can be pacing, defibrillation, or transcutaneous electrical nerve stimulation (TENS) pulses.

[0191] Medical device 700 may include one or more of the following: garment 710; one or more sensing electrodes 712 (e.g., ECG electrodes); one or more therapy electrodes 714; medical device controller 720; connection compartment 730; patient interface compartment 740; strap 750; or any combination of these components. In some examples, at least some of the components of medical device 700 may be configured to be attached to garment 710 that can be worn on the torso of a patient (or, in some examples, permanently integrated into garment 710).

[0192] The medical device controller 720 is operably connected to the sensing electrodes 712, which can be attached to the garment 710 (e.g., using hook and loop fasteners), such as assembled into the garment 710 or removably attached to the garment. In some implementations, the sensing electrodes 712 can be permanently integrated into the garment 710. The medical device controller 720 is operably connected to the therapy electrodes 714. For example, the therapy electrodes 714 can also be assembled into the garment 710, or in some implementations, the therapy electrodes 714 can be permanently integrated into the garment 710.

[0193] remove Figure 7 Assembly configurations other than those shown are possible. For example, the sensing electrode 712 may be configured to be attached to various locations on the body of the patient 702. The sensing electrode 712 may be operably connected to the medical device controller 720 via the connection compartment 730. In some implementations, the sensing electrode 712 may be adhesively attached to the patient 702. In some implementations, the sensing electrode 712 and the treatment electrode 714 may be included on a single integrated patch and applied to the patient's body in an adhesive manner.

[0194] The sensing electrodes 712 may be configured to detect one or more cardiac signals. Examples of these signals include ECG signals, chest wall motion, and / or other sensed cardiac physiological signals from the patient. The sensing electrodes 712 may also be configured to detect other types of patient physiological parameters, such as tissue fluid levels, lung sounds, breath sounds, patient movement, etc. For example, as described in U.S. Pat. No. 6,253,099, entitled “Cardiac Monitoring Electrode Apparatus and Method,” issued on June 26, 2001 (the contents of which are incorporated herein by reference), an example sensing electrode 712 includes a metal electrode with an oxide coating, such as a tantalum pentoxide electrode.

[0195] In some examples, the treatment electrode 714 may also be configured to include a sensor configured to detect ECG signals and other physiological signals of the patient. In some examples, the connection cabin 730 may include a signal processor configured to amplify, filter and digitize the cardiac signals before sending them to the medical device controller 720. One or more treatment electrodes 714 may be configured to perform the treatment when the medical device 700 determines that the treatment of one or more therapeutic defibrillation shocks is ensured to be delivered to the body of the patient 702 based on the signal detected by the sensing electrode 712 and processed by the medical device controller 720. Example treatment electrodes 714 may include conductive metal electrodes such as stainless steel electrodes, wherein in a specific implementation, these conductive metal electrodes include one or more conductive gel deployment devices configured to deliver conductive gel to the metal electrode before delivering the treatment electrode. In some implementations, the medical device as described herein may be configured to switch between a treatment medical device and a monitoring treatment device configured to monitor the patient only (e.g., not provide or perform any treatment function). For example, therapeutic components such as treatment electrodes 714 and associated circuitry can be optionally disconnected from (or connected to) a medical device, or switched away from (or to) a medical device. For example, a medical device can have an optional therapeutic element (e.g., defibrillation and / or pacing electrodes, components, and associated circuitry) configured to operate in a therapeutic mode. The optional therapeutic element can be physically disconnected from the medical device as a means of converting the therapeutic medical device into a monitoring medical device for a specific use (e.g., for operating in a monitoring-only mode) or patient. Optionally, the optional therapeutic element can be deactivated (e.g., using a physical or software switch), thereby essentially rendering the therapeutic medical device as a monitoring medical device for a specific physiological purpose or for a specific patient. As an example of a software switch, an authorized person can access a protected user interface of a medical device and select a preconfigured option or perform some other user action via the interface to deactivate a therapeutic element of the medical device.

[0196] WMD / WCD Controller Description

[0197] Figure 8 A view showing a sample component level of the medical device controller 720 is shown. Figure 8 As shown, the medical device controller 720 may include a therapy delivery circuit 802, a data storage 804, a network interface 806, a user interface 808, at least one battery 810, a sensor interface 812, an alarm manager 814, and at least one processor 818. The patient monitoring medical device may include a medical device controller 720 including the same components as described above, but without the therapy delivery circuit 802 (shown in dashed lines).

[0198] The therapy delivery circuit 802 may be connected to one or more electrodes 820 (e.g., as described above with respect to Figure 7 For example, the therapy delivery circuit 802 may include or be operably connected to circuit components configured to generate and provide a therapeutic shock. These circuit components may include, for example, resistors, capacitors, relays and / or switches, bridges such as H-bridges (e.g., including multiple insulated gate bipolar transistors or IGBTs), voltage and / or current measurement components, and other similar circuit components, which are arranged and connected so that they cooperate with the therapy delivery circuit and operate under the control of one or more processors (e.g., processor 818) to provide, for example, one or more pacing or defibrillation therapy pulses.

[0199] Pacing pulses may be used to treat arrhythmias such as bradycardia (e.g., less than 30 beats per minute) and tachycardia (e.g., greater than 150 beats per minute), for example, using fixed-rate pacing, demand pacing, and anti-tachycardia pacing, etc. Defibrillation pulses may be used to treat ventricular tachycardia and / or ventricular fibrillation.

[0200] The capacitor may include a capacitor bank containing multiple capacitors (e.g., two, three, four or more capacitors) connected in parallel. During the discharge period used for the defibrillation pulse, these capacitors can be switched to a series connection. For example, four capacitors of about 650 microfarads can be used. The surge rating of these capacitors can be between 350 and 500 volts and can be charged from the battery pack in about 15 to 30 seconds.

[0201] For example, each defibrillation pulse can deliver 60 to 180 joules of energy. In some implementations, the defibrillation pulse can be a biphasic truncated exponential waveform, whereby the signal can switch between a positive portion and a negative portion (e.g., a charging direction). Such a waveform can be effective in defibrillating a patient at a lower energy level than other types of defibrillation pulses (e.g., monophasic pulses, etc.). For example, the amplitude and width of the two phases of the energy waveform can be automatically adjusted to deliver a precise amount of energy (e.g., 150 joules) regardless of the patient's body impedance. The therapy delivery circuit 802 can be configured to perform switching and pulse delivery operations, for example, under the control of a processor 818. When delivering energy to a patient, the amount of energy being delivered can be tracked. For example, even when the pulse waveform is dynamically controlled based on factors such as the body impedance of the patient to which the pulse is being delivered, the amount of energy can be maintained at a predetermined constant value.

[0202] The data storage 804 may include one or more non-transitory computer-readable media, such as flash memory, solid-state memory, magnetic memory, optical memory, cache memory, combinations thereof, etc. The data storage 804 may be configured to store executable instructions and data used by the operation of the medical device controller 720. In a particular implementation, the data storage may include executable instructions that, when executed, are configured to cause the processor 818 to perform one or more functions.

[0203] In some examples, the network interface 806 can enable information communication between the medical device controller 720 and one or more other devices or entities via a communication network. For example, in the case where the medical device controller 720 is included in a mobile medical device (such as the medical device 700), the network interface 806 can be configured to communicate with a remote computing device such as a remote server or other similar computing device.

[0204] In a specific implementation, the user interface 808 may include one or more physical interface devices such as input devices, output devices, and combined input / output devices, and a software stack configured to drive the operation of these devices. These user interface elements can render visual, audio, and / or tactile content, including content related to position-specific processing. Thus, the user interface 808 can receive input or provide output, thereby enabling a user to interact with the medical device controller 720.

[0205] The medical device controller 720 may also include at least one battery 810, wherein the battery 810 is configured to provide power to one or more components integrated in the medical device controller 720. The battery 810 may include a rechargeable multi-cell battery pack. In one example implementation, the battery 810 may include more than three 2200 mAh lithium-ion cells that provide power to other device components within the medical device controller 720. For example, the battery 810 may provide its power output in an output range of 20 mA to 1000 mA (e.g., 40 mA), and may support 24 hours, 48 ​​hours, 72 hours, or more hours of operating time between charges. In a specific implementation, the battery capacity, operating time, and type (e.g., lithium ion, nickel cadmium, or nickel metal hydride) may be varied to best suit the specific application of the medical device controller 720.

[0206] The sensor interface 812 can be connected to one or more sensors configured to monitor one or more physiological parameters of the patient. As shown, the sensors can be connected to the medical device controller 720 via a wired or wireless connection. These sensors can include one or more electrocardiogram (ECG) electrodes 822 (e.g., as described above with respect to Figure 7 The invention also includes a plurality of sensors 822 and 823 that are similar to the sensing electrodes 712 described above), other physiological or motion sensors 824, and an interstitial fluid monitor 826 (eg, based on an ultra-wideband radio frequency device).

[0207] The ECG electrodes 822 can monitor the patient's ECG information. For example, the ECG electrodes 822 can be conductive and / or capacitive electrodes configured to measure changes in the patient's electrophysiological phenomena to measure the patient's ECG information. The ECG electrodes 822 can send information describing the ECG signal to the sensor interface 812 for subsequent analysis using algorithms such as the above algorithms. This analysis can be performed by the processor 818 or another processor associated with the system.

[0208] Interstitial fluid monitor 826 can use the technology based on radio frequency (RF) to assess the fluid level and accumulation in the patient's body tissue.For example, interstitial fluid monitor 826 can be configured to measure the intrapulmonary fluid content, is generally used for the diagnosis and follow-up of pulmonary edema or pulmonary congestion in heart failure patients.Interstitial fluid monitor 826 may include one or more antennas, it is configured to guide RF waves through the patient's tissue, and measures the output RF signal in response to the wave through the tissue.In a specific implementation, the output RF signal includes the parameter representing the fluid level in the patient's tissue.Interstitial fluid monitor 826 can send the information describing the interstitial fluid level to sensor interface 812 for subsequent analysis.

[0209] The sensor interface 812 may be connected to any one or a combination of sensing electrodes / other sensors to receive other patient data representative of patient parameters. Once the sensor interface 812 receives the data from the sensor, the data may be directed by the processor 818 to the appropriate components within the medical device controller 720. For example, if cardiac data is collected using the sensor 824 and sent to the sensor interface 812, the sensor interface 812 may send the data to the processor 818, which relays the data to the cardiac event detector. The cardiac event data may also be stored on the data storage 804.

[0210] In a specific implementation, the alarm manager 814 may be configured to manage alarm profiles and notify the expected recipients of events specified in the alarm profiles as being of interest to one or more expected recipients. These expected recipients may include external entities such as users (patients, physicians, or monitoring personnel) and computer systems (monitoring systems or emergency response systems). The alarm manager 814 may be implemented using hardware or a combination of hardware and software. For example, in some examples, the alarm manager 814 may be implemented as a software component stored in the data memory 804 and executed by the processor 818. In this example, the instructions included in the alarm manager 814 may enable the processor 818 to configure the alarm profiles and use these alarm profiles to notify the expected recipients. In other examples, the alarm manager 814 may be an application specific integrated circuit (ASIC) connected to the processor 818 and configured to manage the alarm profiles and use the alarms specified in the alarm profiles to notify the expected recipients. Thus, the examples of the alarm manager 814 are not limited to specific hardware or software implementations.

[0211] In some implementations, the processor 818 includes one or more processors (or one or more processor cores), each of which is configured to perform a series of instructions, wherein the series of instructions generates manipulation data and / or controls the operation of other components of the medical device controller 720. In some implementations, in the case of performing a specific process (e.g., cardiac monitoring), the processor 818 may be configured to perform a specific logic-based judgment based on the received input data, and may also be configured to provide one or more outputs that can be used to control or otherwise inform subsequent processing to be performed by the processor 818 and / or other processors or circuits to which the processor 818 is communicatively connected. Thus, the processor 818 reacts to a specific input stimulus in a specific manner and generates a corresponding output based on the input stimulus. In some example cases, the processor 818 can perform a series of logic transitions, wherein various internal register states and / or states of other bit cells located inside or outside the processor 818 can be set to logic high or logic low. As described herein, the processor 818 can be configured to perform a function in which software is stored in a data memory connected to the processor 818, wherein the software is configured to cause the processor 818 to make a series of various logical decisions that cause the function to be performed. The various components described herein as executable by the processor 818 can be implemented in various forms of dedicated hardware, software, or a combination thereof. For example, the processor can be a digital signal processor (DSP), such as a 24-bit DSP processor, etc. The processor can be, for example, a multi-core processor with two or more processing cores. The processor can be an advanced RISC machine (ARM) processor, such as a 32-bit ARM processor, etc. The processor can execute an embedded operating system and include services provided by the operating system that can be used for file system manipulation, display and audio generation, basic networking, firewalls, data encryption, and communication.

[0212] The specific techniques described herein may be facilitated by the use of a computer-implemented medical device such as a defibrillator that includes computing capabilities. Such a defibrillator or other device may include, for example, Fig. 9The computer system 900 shown in the figure can be communicated with and / or incorporated into the computer system 900 to perform the above operations. For example, one or more of the ECG analyzer module 905, the TTI module 908, the CPR feedback module 910, the PTI module 912 and the logic module 918 can include at least a portion of the computing system 900. The system 900 can be implemented using various forms of digital computers, including computerized defibrillator laptop computers, personal digital assistants, tablet computers and other suitable computers. In addition, the system can include portable storage media such as universal serial bus (USB) flash drives, etc. For example, a USB flash drive can store operating systems and other applications. A USB flash drive can include input / output components, such as a wireless transmitter or a USB connector that can be inserted into a USB port of another computing device, etc.

[0213] The system 900 includes a processor 910, a memory 920, a storage device 930, and an input / output device 940. The components 910, 920, 930, and 940 are interconnected using a system bus 950. The processor 910 is capable of processing instructions for execution within the system 900. The processor may be designed using any of a plurality of architectures. For example, the processor 910 may be a CISC (Complex Instruction Set Computer) processor, a RISC (Reduced Instruction Set Computer) processor, or a MISC (Minimum Instruction Set Computer) processor.

[0214] In one implementation, the processor 910 is a single-threaded processor. In another implementation, the processor 910 is a multi-threaded processor. The processor 910 is capable of processing instructions stored in the memory 920 or on the storage device 930 to display graphical information used by the user interface on the input / output device 940.

[0215] The memory 920 stores information within the system 900. In one implementation, the memory 920 is a computer-readable medium. In one implementation, the memory 920 is a volatile memory unit. In another implementation, the memory 920 is a non-volatile memory unit.

[0216] The storage device 930 can provide mass storage for the system 900. In one implementation, the storage device 930 is a computer-readable medium. In various implementations, the storage device 930 can be a floppy disk device, a hard disk device, an optical disk device, or a tape device.

[0217] The input / output device 940 provides input / output operations to the system 900. In one implementation, the input / output device 940 includes a keyboard and / or a pointer device. In another implementation, the input / output device 940 includes a display unit for displaying a graphical user interface.

[0218] The features described can be implemented in digital electronic circuits, in computer hardware, firmware, software, or in a combination thereof. The device can be implemented in a computer program product tangibly embodied in an information carrier (e.g., a machine-readable storage medium for execution by a programmable processor), and the method steps can be performed by: the programmable processor executes a program of instructions to perform the functions described by operating on input data and generating outputs. The features described can be advantageously implemented in one or more computer programs executable on a programmable system including at least one programmable processor, wherein the at least one programmable processor is connected to receive and send data and instructions relative to a data storage system, at least one input device, and at least one output device. A computer program is a set of instructions that can be used directly or indirectly in a computer to perform a specific or produce a specific result. A computer program can be written in any form of programming language including a compiled language or an interpreted language, and can be deployed in any form, including being deployed as a standalone program, or as a module, component, subroutine, or other unit suitable for use in a computing environment.

[0219] For example, suitable processors for executing a program of instructions include both general-purpose microprocessors and special-purpose microprocessors, and a sole processor or one of multiple processors of any kind of computer. Generally, a computer will receive instructions and data from a read-only memory or a random access memory or both. The basic elements of a computer are a processor for executing instructions and one or more memories for storing instructions and data. Typically, a computer will also include one or more mass storage devices for storing data files, or be operably connected to communicate with one or more mass storage devices; these devices include disks (such as internal hard disks and removable disks, etc.), magneto-optical disks, and optical disks. Storage devices suitable for tangibly embodying computer program instructions and data include all forms of non-volatile memory, such as semiconductor memory devices (such as EPROM, EEPROM, etc.) and flash memory devices, disks such as internal hard disks and removable disks, magneto-optical disks, and CD-ROM disks and DVD-ROM disks. The processor and memory can be supplemented by or incorporated into an ASIC (Application Specific Integrated Circuit).

[0220] To provide interaction with a user, these features may be implemented on a computer having an LCD (liquid crystal display) or LED display for displaying information to the user, and a keyboard and pointing device such as a mouse or trackball by which the user can provide input to the computer.

[0221] These features can be implemented in a computer system that includes a back-end component (such as a data server, etc.), or an intermediate component (such as an application server or an Internet server, etc.), or a front-end component (such as a client computer with a graphical user interface or an Internet browser, etc.), or any combination thereof. The components of the system can be connected by any form or medium of digital data communication (such as a communication network, etc.). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), a peer-to-peer network (with temporary or static members), a grid computing architecture, and the Internet.

[0222] A computer system may include clients and servers. Clients and servers are generally remote from each other and typically interact via a network such as described above. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other.

[0223] Although the subject matter contained herein has been described in detail for the purpose of illustration, it should be understood that such detail is for this purpose only and that the invention is not limited to the disclosed implementations, but rather is intended to cover modifications and equivalent arrangements within the spirit and scope of the appended claims. For example, it should be understood that the invention contemplates that, to the extent possible, one or more features of any implementation may be combined with one or more features of any other implementation.

[0224] Other examples are within the scope and spirit of the specification and claims. In addition, the above-mentioned specific functions can be implemented using software, hardware, firmware, hard wiring, or any combination of these. The features that implement the functions can also be physically located in various locations, including being distributed as part of the functions and implemented at different physical locations.

[0225] Other implementations than those described above may be employed and are intended to be covered by the accompanying claims.

[0226] Priority Claim

[0227] This application claims priority to U.S. Patent Application Serial No. 62 / 279,713, filed January 16, 2016, the entire contents of which are incorporated herein by reference.

Claims

1. A defibrillator, comprising: one or more capacitors for delivering a defibrillation shock to a patient; one or more electronic ports for receiving signals from at least one sensor for obtaining an electrocardiogram (ECG) of the patient; Patient treatment circuitry executable on one or more processing devices configured to perform operations including: receiving data representing an ECG signal, detecting delivery of chest compressions to the patient, processing an ECG signal during delivery of the chest compressions to generate a first decision whether an ECG rhythm represented by the ECG signal is shockable or non-shockable, detecting a pause in the delivery of the chest compressions to the patient, processing the ECG signal during the pause in the delivery of the chest compressions to generate a second decision whether the ECG rhythm represented by the ECG signal is shockable or non-shockable, determining whether the first decision and the second decision are consistent, and generating a final decision of whether the ECG rhythm is shockable or non-shockable based on the determination of whether the first decision and the second decision are consistent; as well as An output circuit is configured to provide an output based on the final decision.

2. The defibrillator according to claim 1, wherein: Processing the ECG signal during the pause in delivery of chest compressions to generate a second determination whether the ECG rhythm represented by the ECG signal is shockable or non-shockable is initiated within a predetermined stabilization period after the pause in delivery of chest compressions to the patient begins.

3. The defibrillator according to claim 2, wherein: The predetermined stabilization time period is at least 1.5 seconds.

4. The defibrillator of claim 1 , further comprising a compression port for receiving a motion signal from a motion sensor for obtaining an indication of delivery of the chest compressions to the patient and an indication of a pause in delivery of the chest compressions to the patient.

5. The defibrillator according to claim 4, wherein: The motion sensor includes an accelerometer, a velocity sensor or a displacement sensor.

6. The defibrillator according to claim 1, wherein: The at least one sensor includes electrode pads for delivering a defibrillation shock to the patient.

7. The defibrillator according to claim 6, wherein: The electrode pads are used to provide an indication of the delivery of the chest compressions to the patient and an indication of a pause in the delivery of the chest compressions to the patient.

8. The defibrillator according to claim 1, wherein: Processing the ECG signal during the delivery of the chest compressions to generate the first decision is based on a first processing algorithm that includes compression artifact removal.

9. The defibrillator according to claim 8, wherein: The first processing algorithm includes a frequency-based analysis.

10. The defibrillator according to claim 8, wherein: The operations also include: After generating the first determination whether the ECG rhythm represented by the ECG signal is shockable or non-shockable, an initial prompt is generated to pause delivery of the chest compressions to the patient.

11. The defibrillator according to claim 10, wherein: The operations also include: It is determined that delivery of the chest compressions to the patient continues after the initial prompt to pause the delivery of the chest compressions is provided.

12. The defibrillator according to claim 11, wherein: The operations also include: Processing of the ECG signal during the delivery of the chest compressions continues based on the first processing algorithm.

13. The defibrillator according to claim 11, wherein: The operations also include: After the initial prompt and after it is determined that delivery of the chest compressions to the patient is continuing, a subsequent prompt is generated to pause delivery of the chest compressions to the patient.

14. The defibrillator according to claim 13, wherein: The operations also include: detecting a pause in the delivery of the chest compressions to the patient during a time period following the subsequent prompt, and In response to identifying a pause in the delivery of chest compressions to the patient, processing the ECG signal during the pause in the delivery of chest compressions begins to generate a second decision at the beginning of a next time segment.

15. The defibrillator according to claim 10, wherein: The first decision determined based on the first processing algorithm has a first level of accuracy that is lower than a second level of accuracy of the second decision.

16. The defibrillator according to claim 15, wherein: The first level of accuracy and the second level of accuracy each represent a positive predictive value or a negative predictive value.

17. The defibrillator according to claim 16, wherein: The first level of accuracy and the second level of accuracy each represent a threshold value for combined sensitivity and specificity.

18. The defibrillator according to claim 1, wherein: The one or more processing devices are configured to initiate charging of the one or more capacitors if the patient is in a shockable state.

19. The defibrillator according to claim 1, wherein: The defibrillator includes at least one of the following: an automatic external defibrillator, and a defibrillator monitor.

20. The defibrillator of claim 1, wherein: The output circuit is configured to present a user interface configured to provide feedback indicating whether the patient is in a shockable state.

21. The defibrillator of claim 1, wherein: The operations also include: Variable length time segments of the ECG signal during the time period when chest compressions are paused are processed to determine a second decision whether the ECG rhythm represented by the ECG signal is shockable or non-shockable.

22. The defibrillator according to claim 21, wherein: The accuracy of determining that the patient is in a shockable state increases as the duration of the variable-length time segment increases.

23. The defibrillator of claim 1, wherein: The one or more processing devices are configured to access a set of one or more clauses representing constraints on one or more features, the one or more features each representing a characteristic of a waveform of the ECG signal.

24. The defibrillator of claim 23, wherein: The clauses are each associated with a respective threshold accuracy level, and the one or more processing devices are configured to evaluate at least one of the clauses with respect to the ECG signal to determine whether the ECG signal meets the threshold accuracy level of the respective clause.

25. The defibrillator of claim 1, further comprising a display for indicating whether the ECG rhythm is shockable or non-shockable based on the output of the final decision.

26. The defibrillator of claim 25, the operations further comprising: A defibrillation shock is delivered to the patient based on an output of the final decision.

27. The defibrillator of claim 1, wherein: Processing of ECG signals during delivery of the chest compressions is performed over a plurality of CPR segments to generate the first decision.

28. The defibrillator of claim 1, wherein: Processing of the ECG signal during pauses in the delivery of the chest compressions is performed over a plurality of no-CPR segments to generate the second decision.

29. The defibrillator of claim 1, wherein: Processing the ECG signal during the pause in the delivery of chest compressions to generate the second decision includes processing variable length time segments of the ECG signal.

30. The defibrillator of claim 29, wherein: Processing the variable length time segments of the ECG signal comprises: identifying one or more features of the variable length time segments based on an initial set of rules corresponding to an initial time period to determine whether the ECG rhythm represented by the ECG signal is shockable or non-shockable, Based on the initial rule set, determining whether the ECG rhythm represented by the ECG signal is shockable or non-shockable, or determining whether to extend the variable length time segment for further analysis, extending said initial time period of said variable length time segment with one or more additional time periods, and One or more features of the variable length time segments are identified based on an adjusted rule set corresponding to the adjusted time period to update a determination of whether the ECG rhythm is shockable or non-shockable.

Images