System and method for electrocardiogram radial depth navigation of intracardiac devices

Abstract

Various systems and methods are provided for electrocardiogram radial depth navigation to insert an intracardiac device into the heart. In one example, a method includes, during insertion of an intracardiac device into a muscle or chamber of a heart, acquiring an intracardiac electrogram signal by an electrode of the intracardiac device, and outputting a radial depth indication relative to a myocardium of the heart based on electronic processing of the intracardiac electrogram signal.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application is an international application claiming priority to U.S. Provisional Application No. 63 / 371,490, filed on August 15, 2022, and entitled “SYSTEM AND METHOD FOR ELECTROCARDIOCARDIAL RADIAL DEPTH NAVIGATION OF INTRACARDIAC DEVICES.” The entire contents of the above-referenced applications are hereby incorporated by reference for all purposes. Technical Field

[0002] The present specification relates generally to catheter-based cardiac surgery, and electrocardiogram-based methods of navigating catheter-based cardiac devices during surgery. Background Art

[0003] Some structural heart surgeries use catheter-based cardiac devices that can be navigated to a target location within the heart muscle via a freely steerable guidewire. As an example, a procedure known as myocardial intramural remodeling by transvenous tethering (e.g., MIRTH) involves implanting an intramyocardial tension element within the ventricular wall to narrow the left ventricle, thereby providing treatment for heart failure. As another example, endocardial midline septotomy (e.g., SESAME) is a transcatheter myotomy that can be used to relieve left ventricular outflow tract obstruction. As yet another example, cerclage valvuloplasty is a procedure that uses a catheter to reduce mitral valve regurgitation. As yet another example, ablation treatment of ventricular arrhythmias may require the use of local electrocardiographic measurements from a catheter or guidewire to locate the arrhythmia propagation pathway within the myocardium, and deliver ablative energy or agents to eliminate the intramyocardial portion of the arrhythmia propagation pathway.

[0004] In each of these surgeries, a guidewire may be steered to a target location to assist in implanting a catheter-based cardiac device or delivering ablative energy. Alternatively, a catheter-based cardiac device may be steered directly without the use of a guidewire. During guidewire / device steering, various imaging techniques may be used to provide navigation feedback, such as x-ray fluoroscopy and echocardiography. However, x-ray fluoroscopy cannot visualize soft tissue, which causes the guidewire or device to appear to be floating freely in space. As another example, echocardiography is hindered by off-axis imaging planes, which may cause the guidewire or device to appear deeper or shallower than where it is actually located. Therefore, when using traditional imaging modes for intramyocardial device navigation, it may be difficult to determine the radial position of the guidewire or device between the endocardial and epicardial surfaces of the myocardium. For the purposes of this specification, the term "intramyocardial" is used interchangeably with the term "intracardiac", even though the former refers to a location within the heart muscle and the latter refers to a location within and outside the heart muscle.

[0005] However, the radial depth of device placement may affect the outcome of the procedure. For example, it may be desirable to implant an intramyocardial tension element used in a MIRTH procedure at a radial depth equidistant from the endocardium and epicardium (e.g., in the mid-myocardium). As another example, it may be desirable to implant a SESAME implant deep enough in the myocardium to relieve left ventricular outflow tract obstruction, but shallow enough to prevent the creation of a ventricular septal defect. As still another example, by knowing the location of the exit point of the cerclage catheter, the occurrence of heart block may be reduced. As still another example, it may be desirable to navigate an ablation device within the mid-myocardium to reach ablation targets that are otherwise inaccessible with traditional endocardial or epicardial catheter ablation. Summary of the invention

[0006] In one example, the problem described above can be solved by a method comprising acquiring intracardiac electrogram signals through electrodes of an intracardiac device (such as a guidewire or ablation catheter) during insertion of an intracardiac device into cardiac muscle and associated chambers, and outputting a radial depth indication relative to the myocardium of the heart based on electronic processing of the intracardiac electrogram signals. In this way, radial depth navigation feedback can be provided during insertion of the intracardiac device, thereby, for example, more efficiently and accurately placing a catheter-based cardiac device.

[0007] It should be understood that the above brief description is provided to introduce in simplified form a series of concepts that are further described in the detailed description. It is not meant to identify key or essential features of the claimed subject matter, the scope of which is uniquely defined by the claims that follow the detailed description. Furthermore, the claimed subject matter is not limited to embodiments that address any disadvantages noted above or in any part of the disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] The present disclosure will be better understood from the following description of non-limiting embodiments with reference to the accompanying drawings, in which: Figure 1 An embodiment of an electrocardiogram radial depth navigation system is described; Figure 2 shows a high-level block diagram of an electrocardiographic radial depth navigation algorithm that can be used to provide real-time radial depth navigation feedback during insertion of an intracardiac device into the heart; Figure 3 A block diagram showing a workflow that can be used to train an intramyocardial electrogram classifier; Figure 4 Example monopolar electrode data relative to radial depth of the myocardium is shown; Figure 5 shows example waveform characteristics that may be extracted from an intracardiac electrogram signal measured by a monopolar electrode positioned at a first radial depth; Figure 6 shows example waveform characteristics that may be extracted from an intracardiac electrogram signal measured by a monopolar electrode positioned at a second radial depth; Figure 7 shows example waveform characteristics that may be extracted from an intracardiac electrogram signal measured by a monopolar electrode positioned at a third radial depth; Figure 8 shows example waveform characteristics that may be extracted from an intramyocardial electrogram signal measured by a monopolar electrode positioned at a fourth radial depth; Fig. 9 A table summarizing the qualitative characteristics of intracardiac electrograms at different radial depth categories is shown; Fig.10 showing time-based changes in the mid-level electrocardiogram from a non-moving monopolar electrode; and Fig.11 A flow chart of an example method for providing real-time radial depth navigation feedback during insertion of an intracardiac device into a heart is shown. DETAILED DESCRIPTION

[0009] The following description relates to systems and methods for navigating intracardiac devices by electrocardiogram radial depth navigation (also referred to herein as EDEN). Examples of intracardiac devices that can be navigated by EDEN as described herein include catheters, intramyocardial tension elements, electrophysiological mapping and ablation catheters, myocardial biopsy devices, myocardial drug delivery catheters or needles, myocardial electrical stimulation devices, and / or myocardial pacing electrodes. In some instances, one or more intracardiac devices can be navigated by a guidewire, and thus the guidewire can also be an example of an intracardiac device that can be navigated by EDEN. Although real-time imaging techniques such as x-ray fluoroscopy can provide information about the longitudinal (e.g., base to apex) and circumferential (e.g., "clock face") position of an intracardiac device in the heart, these techniques lack information about the radial position (also referred to as radial depth) of the intracardiac device in the myocardium. As used herein with respect to the heart, the terms "radial position" and "radial depth" represent the relative position of the heart between the endocardial surface and the epicardial surface in a single dimension. As will be further described herein, the radial depth of the intracardiac device can be classified according to its relative position between the endocardial border and the epicardial border (or beyond said border) by a depth navigation algorithm (which can be a machine learning or deep learning based algorithm or a logic based algorithm) that uses intramyocardial electrograms measured by electrodes of the intracardiac device itself. Thus, the depth navigation algorithm can provide real-time feedback to the operator of the intracardiac device during insertion, which can improve the efficiency and accuracy of the tip position of the intracardiac device (e.g., a guidewire).

[0010] Figure 1An embodiment of an EDEN system for intramyocardial device navigation is shown. The EDEN system may include a deep navigation algorithm such as Figure 2 The depth navigation algorithm is configured to provide real-time feedback on the radial depth position of an intracardiac device based on electrograms acquired through electrodes of the intracardiac device. For example, the depth navigation algorithm may include a trained machine learning or deep learning based classifier or a logic based classifier that may determine the radial depth category of each acquired electrogram. For example, the classifier may be based on Figure 3 The example scheme shown in . Figure 4 As shown in , different depths between the endocardial and epicardial boundaries of the myocardium and outside the myocardium will produce different characteristic waveform features. These waveform features can be extracted, and examples of the waveform features are shown in Figure 5-8 and is used by the classifier to distinguish different radial depth categories. Fig. 9 Qualitative differences between waveform characteristics for different radial depth categories are summarized. In addition, electrograms acquired with electrodes in intracardiac devices may vary over time in a predictable manner, as in the case where the electrodes are held in place rather than being steered through the myocardium. Fig.10 Thus, the depth navigation algorithm can take electrode movement (or lack thereof) into account based on known changes in characteristic waveform features. For example, characteristic features of consecutive electrograms can be stored in a memory. If the change in these features is less than a predefined threshold, the electrode can be considered to remain at the same radial depth. After the first condition is met, the change in the characteristic features in the time domain can be compared with the Fig.10 The depth navigation algorithm can be used during insertion to compare the resting electrode signal trend in the depth navigation algorithm to confirm the resting position of the electrode. Fig.11 The method provides real-time intracardiac device navigation feedback.

[0011] Now turning to the attached figure, Figure 1 An example of an electrocardiographic radial depth navigation (EDEN) system 100 is schematically depicted relative to a patient's heart 101. The heart 101 is shown as a simplified short-axis view depicting a cross-section of a left ventricle 102, a right ventricle 106, an interventricular septum (IVS) 109 separating the left ventricle 102 and the right ventricle 106, and a coronary sinus 108. The heart 101 includes a myocardium 104 that is bounded externally by an epicardium 103 and internally by an endocardium 105. Thus, the myocardium 104 has a radial depth 107 from the epicardium 103 to the endocardium 105. Locations external to the epicardium 103 may be referred to as extra-epicardial or pericardial, and locations internal to the endocardium 105 may be referred to as intracavitary or intra-atrial.

[0012] The EDEN system 100 includes an exposed conductor 112 positioned on a guidewire 110. The exposed conductor 112 includes an intracardiac sensing electrode of the EDEN system 100, and therefore may also be referred to herein as an electrode. During guidewire navigation, at least a portion of the guidewire 110 is inserted into the patient, and the remainder of the guidewire 110 may remain outside the patient to enable guidewire manipulation and electrogram recording, as will be described in detail below. In some instances (such as the illustrated instances), the guidewire 110 may be fed through a sheath 114 (and / or a catheter and / or a microcatheter) connected to a hub 116. The sheath 114 may be a flexible material that is inserted into the patient's blood vessel to provide support and controlled access for the insertion of the guidewire 110. For example, the sheath 114 may be anchored to the outside of the patient by a hub 116. The guidewire 110 may also be inserted through a torque device 118 (also referred to as a torquer), which may be used to manipulate the guidewire 110. For example, the torque device 118 can help grasp and steer the guidewire 110. Figure 1 While described herein with respect to a guidewire with electrodes, it should be understood that the EDEN system 100 may instead include another suitable intracardiac device, such as an ablation catheter, a myocardial biopsy device, a myocardial drug delivery catheter or needle, a myocardial electrical stimulation or contractility modulation device, or a myocardial pacing electrode. The intracardiac device may include one or more intracardiac sensing electrodes that may be used to obtain an electrogram that may be processed as described herein to determine the radial depth of the intracardiac device.

[0013] The guidewire 110 comprises a thin (e.g., outer diameter in the range of 0.01 inches to 0.04 inches) cylindrical material having a stiffness (or flexibility) that enables insertion into the myocardium 104 and navigation within the myocardium. For example, the stiffness of the guidewire 110 can be greater than the stiffness used for the vascular subselection procedure. In addition, the tip of the guidewire 110 can be shaped to facilitate guidewire navigation by twisting and advancing. In addition, the guidewire 110 can include a conductive transmission wire, such as a copper wire, which is wrapped in a stainless steel and / or nickel titanium alloy (e.g., Nitinol) braided or coiled wire to improve pushability and kink resistance. Further, except for the exposed conductor 112, the length of the guidewire 110 can be electrically insulated. For example, except for the exposed conductor 112 and the connection point that electrically couples the exposed conductor 112 to the signal processor 122 through the conductive transmission line, the guidewire 110 can be coated with one or more layers of insulation. In this way, the guide wire 110 can include insulating regions and non-insulating regions (e.g., exposed conductors). In some instances, the electrodes of the intracardiac device (e.g., exposed conductors 112) can be monopolar electrodes. In other instances, the electrodes can be bipolar or multipolar. For example, the guide wire can incorporate multiple electrically isolated insulator-conductor subassemblies for connecting multiple electrical channels to independent or multiplexed transmission line systems. In some dual conductor configurations, this will enable the collection of local bipolar electrograms. For example, the inclusion of monopolar electrodes can reduce the cost and complexity of manufacturing the EDEN system 100 relative to bipolar or multipolar electrode configurations, which can generate electrograms with higher signal-to-noise ratios.

[0014] In some examples, the guidewire 110 can be concentric with a catheter (not shown) to be implanted. For example, the outer diameter of the guidewire 110 can be smaller than the inner diameter of the catheter, and the outer diameter of the catheter can be smaller than the inner diameter of the sheath 114. After the guidewire 110 is navigated to the target location in the heart 101, the catheter can be advanced through the sheath 114 and over the guidewire 110 to position the catheter at the target location.

[0015] In some examples, the exposed conductor 112 is positioned at the distal tip of the guidewire 110, such as Figure 1 In other examples, the exposed conductor 112 can be positioned elsewhere along the length of the guidewire 110. Further, the length of the exposed conductor 112 can be calibrated to narrow electrogram sensing to a focal area of ​​a few millimeters (mm) or less within the myocardium 104 while providing a signal large enough for effective acquisition. In some examples, such as Figure 1In the example shown in , there can be a single exposed conductor, thereby making the exposed conductor 112 a monopolar intramyocardial electrode, while other examples may include multiple exposed conductors, as explained above. The exposed conductor 112 may contain, for example, gold, stainless steel, nickel-titanium alloy, MP35N, 35NLT, silver, platinum, platinum-iridium, and / or conductive carbon.

[0016] The signal processor 122 may further include one or more amplifiers 134, one or more filters 136, an analog-to-digital converter (ADC) 138, a buffer 140, and a power supply 142. The power supply 142 may include a battery, such as a lithium-ion battery. In some examples, the battery may be a rechargeable battery. Additionally or alternatively, the power supply 142 may receive power from a wall outlet (e.g., via a power adapter). The buffer 140 may be a memory component, such as a temporary storage portion of a random access memory, that stores data transmitted between the signal processor 122 and other data processing components (such as the computing device 120 described further below).

[0017] One or more amplifiers 134 may include an electrically isolated or optically isolated amplifier that receives and amplifies the intramyocardial electrogram signal from the exposed conductor 112. One or more filters 136 may include a bandpass filter, a bandstop filter, a high-pass filter, a low-pass filter, or a combination thereof. As an example, one or more filters 136 may include a low-pass filter with a cutoff frequency in the range of 100 Hertz (Hz) to 150 Hz and a high-pass filter with a cutoff frequency in the range of 0.05 Hz to 0.5 Hz. Additionally or alternatively, a bandstop filter may also be used to eliminate mains interference from the power supply 142 that is carried through the intramyocardial electrogram signal. For example, the bandstop filter may filter out frequencies of 50 Hz or 60 Hz, depending on the specifications of the power supply 142.

[0018] The filtered and amplified analog intramyocardial electrogram signals may be converted to digital signals using a sampling frequency in a range between 200 Hz and 2000 Hz by, for example, ADC 138. Alternatively, the intramyocardial electrogram signals may be converted to digital signals by ADC 138 after being amplified by one or more amplifiers 134 and before being filtered by one or more filters 136, such as when one or more filters 136 are digital filters and the filtering process is performed digitally.

[0019] The EDEN system 100 further includes a computing device 120, which can be configured to record and process the digitally processed electrogram signals received from the signal processor 122. For example, the computing device 120 can be a data processor. The computing device 120 can be in wired or wireless electronic communication with the signal processor 122. In other examples, the signal processor 122 can be integrated with the computing device 120, such as being housed in a common housing as a single unit. The time interval data of the digitally processed signal can be transferred to the computing device 120 in the form of a sliding window. For example, the window width of the sliding window can be 0.2 seconds. In other examples, the window width can be adaptively adjusted based on the rhythm of the heart 101. Further, the time interval data can be stored in a buffer 140 before being transmitted to the computing device 120 to avoid data loss. Alternatively, the signal processor 122 can transmit continuous real-time digital signal data to the computing device 120. As used herein, the term "real-time" refers to a process that is performed substantially when it occurs. As another example, the term "real-time" refers to a process that is performed without deliberate delay.

[0020] The computing device 120 includes a processor 124 configured to execute machine-readable instructions stored in a non-transitory memory 126. The processor 124 may be single-core or multi-core, and the program executed by the processor 124 may be configured for parallel processing or distributed processing. In some instances, the processor 124 may optionally include separate components distributed in two or more devices, which may be remotely located and / or may be configured for coordinated processing. In some instances, one or more aspects of the processor 124 may be virtualized and executed by a remotely accessible networked computing device configured in a cloud computing configuration. In some instances, the processor 124 may include other electronic components capable of performing processing functions, such as a digital signal processor, an FPGA, or a graphics board. In some instances, the processor 124 may include multiple electronic components capable of performing processing functions. For example, the processor 124 may include two or more electronic components selected from a plurality of possible electronic components, including a central processing unit, a digital signal processor, a field programmable gate array, and a graphics card. In still other examples, processor 124 may be configured as a graphics processing unit (GPU), which includes a parallel computing architecture and parallel processing capabilities.

[0021] The non-transitory memory 126 may store the digitized processed intramyocardial electrogram signals received from the signal processor 122, and may further store an executable program for analyzing the digitized processed intramyocardial electrogram signals. In some instances, the non-transitory memory 126 may include components located at two or more devices, which may be remotely located and / or may be configured for coordinated processing. In some instances, one or more aspects of the non-transitory memory 126 may include a remotely accessible networked storage device configured in a cloud computing configuration. In some instances, such as Figure 1 As shown, the non-transitory memory 126 stores a depth navigation algorithm 128. The depth navigation algorithm 128 includes one or more algorithms for processing the digitized intramyocardial electrogram signal and determining the radial depth of the intracardiac device electrode (e.g., the electrode of the exposed conductor 112) in the myocardium 104 based on the signal, including a machine learning-based algorithm, a deep learning-based algorithm, and / or a logic-based algorithm. In some instances, the depth navigation algorithm 128 may include a signal processing algorithm that applies a preprogrammed transformation to the digitized intramyocardial electrogram signal. For example, each time interval data segment may be vertically adjusted by comparing the mean value of the signal to zero by curve fitting (e.g., polynomial curve fitting) to compensate for direct current (DC) bias. As an example, each data segment may be fitted to a sixth-order polynomial curve, but lower or higher order polynomials may be used in other instances. Further, the depth navigation algorithm 128 may apply one or more digital filters, such as a zero-phase low-pass filter, such as a sixth-order Butterworth infinite impulse response (IIR) filter, to the vertically adjusted signal. For example, a zero-phase low-pass filter may be applied with a cutoff frequency of 60 Hz. Alternatively, lower-order or higher-order filters may also be applied. Further, the depth navigation algorithm 128 may also apply other types of IIR or finite impulse response (FIR) digital filters.

[0022] The sequentially acquired data segments of the digitized, transformed and filtered intramyocardial electrogram signal may be concatenated and processed by the feature extractor of the deep navigation algorithm 128 using a sliding window. For example, a typical surface electrocardiogram (ECG) waveform may contain a P wave during atrial depolarization, a QRS complex during ventricular depolarization, and a T wave during ventricular repolarization. This may result in five definable points within the waveform: P, Q, R, S, and T. The feature extractor may identify signal features, such as R peaks and ST segment maximums, and may index the detected signal features with timestamps and / or sample numbers. The extracted features may further include features describing the signal and feature extraction quality, such as signal peaks, noise levels, and signal-to-noise ratios (SNRs). The feature extractor may further search for local extrema with prominence in the range of 0.01-50 in the entire concatenated signal. The prominence threshold may be adjusted based on the signal quality and gain of one or more amplifiers 134. The two most recently identified consecutive signal peaks may be stored in the non-transitory memory 126. Further, a signal peak detector can be performed by measuring the slope of the time domain signal using a threshold or moving average technique (e.g., Pan and Tompkins algorithm). As another example, the signal quality can be determined based on the SNR or different ECG components (such as QRS complex or ST segment, etc.). As shown below, Fig.11 As described in detail, low SNR signals may be removed from further processing by the depth navigation algorithm 128 .

[0023] In some instances, the depth navigation algorithm 128 can use the most recent peak-to-peak interval (e.g., RR interval) to calculate the heart rate of the heart 101. Alternatively, the heart rate can also be calculated by averaging multiple previous RR intervals. Further, the depth navigation algorithm 128 can segment the last cardiac cycle (e.g., ECG cycle) by cutting the signal at a first percentage of the peak-to-peak interval before the signal peak (e.g., R peak) and a second remaining percentage of the peak-to-peak interval after the signal peak. For example, the first percentage can be 47% and the second percentage can be 53%. Therefore, the cardiac cycle can start at 47% of the peak-to-peak interval before the signal peak and end at 53% of the peak-to-peak interval after the signal peak. Alternatively, the intramyocardial electrogram signal cardiac cycle segmentation start and end points can be adjusted according to the examples given above (e.g., based on input received from an operator, such as through the user input device 144).

[0024] The feature extractor of the deep navigation algorithm 128 can identify and extract various clear features of the intramyocardial electrogram signal, including TP interval amplitude, P wave amplitude, PQ or PR interval amplitude; Q, R and S wave amplitude; ST segment amplitude; fusion of R wave and ST segment; range of signal amplitude; maximum signal amplitude; minimum signal amplitude; structural similarity between the intramyocardial electrogram waveform and the predefined waveform; variation and standard deviation of the intramyocardial electrogram waveform; etc. In some instances, the intramyocardial electrogram signal can be normalized by making the standard deviation of the signal equal to 1 and the mean value of the signal equal to 0. Further, since the ECG cycle components such as P waves, QRS complexes and T waves are generated by ventricular and atrial repolarization / depolarization, there is a time-locked relationship between these components. Therefore, these components can be segmented using indexing of ECG segments using predefined time constraints. The predefined time constraints can be adjusted by an operator (e.g., through a user input device 144) in response to abnormal ECG activity (such as fibrillation). In some examples, a surface ECG may be obtained by electrodes (not shown) placed on the patient's skin. In such examples, the surface ECG may optionally be used to index intramyocardial electrogram signals.

[0025] In some examples, the deep navigation algorithm 128 may include a classification algorithm based on machine learning or deep learning or based on logic, also referred to herein as a classifier. The classification algorithm may classify the intramyocardial electrogram signal into separate deep categories based on characteristic features extracted by the feature extractor. Examples of deep categories will be described below with respect to Figure 2 and 4 , and the examples of electrogram features that distinguish each depth class will be relative to Fig. 9 For example, the classification algorithm may receive as input the processed intramyocardial electrogram signal (such as the indexed sliding window segments and / or the segmented heartbeat cycles) and output a deep class. The classification algorithm may also receive as input the extracted features associated with each segment. The classification will be described below with respect to Figure 3 and 4 Further described. Further, the depth navigation algorithm 128 may output a position-based indication and / or a category-based depth indication based on the determined depth category to inform the operator of the current radial depth of the intracardiac device (eg, guidewire 110) during insertion.

[0026] Additionally or alternatively, the classification algorithm of the deep navigation algorithm 128 may include one or more deep learning networks, the one or more deep learning networks including a plurality of weights and biases, activation functions, loss functions, gradient descent algorithms, and instructions for implementing the one or more deep learning networks to process input intramyocardial electrogram signals. Additionally or alternatively, the deep navigation algorithm 128 may store instructions for implementing a neural network (such as a convolutional neural network) for classifying processed intramyocardial electrogram segments based on the radial depth of the sensing electrode (e.g., exposed conductor 112) of the intracardiac device in the myocardium 104. The deep navigation algorithm 128 may include trained and / or untrained neural networks, and may further include training routines or parameters (e.g., weights and biases) associated with one or more neural network models stored therein.

[0027] In some examples, the classification algorithm of the deep navigation algorithm 128 can be trained by a training module 132. The training module 132 can include machine executable instructions for training the classification algorithm stored in the deep navigation algorithm 128, and a dataset of individual intramyocardial electrograms labeled by experienced cardiologists. For example, the dataset can include more than 20,000 labeled intramyocardial electrograms. Additional details about training are described below with respect to Figure 3 Give a description.

[0028] In some examples, the sensor 146 can be positioned in the torque device 118. The sensor 146 can be, for example, an accelerometer that provides feedback to the signal processor and / or computing device 120 regarding the movement of the guidewire 110. Fig.10 In detail, intramyocardial electrograms extracted from the same location may evolve in a predictable manner over time due to myocardial cell recovery, and thus the deep navigation algorithm 128 may utilize knowledge of whether the guidewire 110 has moved to classify the electrograms.

[0029] The computing device 120 further includes a user input device 144 and a display device 130. The user input device 144 may include one or more of a touch screen, a keyboard, a mouse, a touch pad, a motion sensing camera, or other devices configured to enable a user to enter data, interact with the data, and / or manipulate the data within the computing device 120. The display device 130 may include one or more display devices utilizing any type of display technology, such as a monitor, a touch screen, and / or a projector. In some instances, the display device 130 may include a computer monitor and may display raw and / or processed electrogram data and depth indications. The display device 130 may be combined with the processor 124, the non-transitory memory 126, and / or the user input device 144 in a common housing, or may be a peripheral device.

[0030] Understandably, Figure 1 The EDEN system 100 shown in is an example embodiment, and other EDEN systems with similar components are possible. For example, another suitable EDEN system may include more, fewer, or different components, such as an ablation catheter, a myocardial biopsy device, a myocardial drug delivery catheter or needle, a myocardial electrical stimulation or modulation device, or a myocardial pacing electrode. The intracardiac device may include one or more intracardiac sensing electrodes, which may be used to obtain an electrogram, which may be processed as described herein to determine the radial depth of the intracardiac device.

[0031] Reference now Figure 2 , shows a high-level block diagram of an EDEN algorithm 200. The EDEN algorithm 200 may be, for example, the depth navigation algorithm 128, and may be implemented by Figure 1 The EDEN algorithm 200 may be implemented in the EDEN system 100 of the embodiment of the present invention. ... Figure 1 The device operates continuously during the heart 101 so that the data is analyzed in essentially real time as it is acquired to provide real-time device depth navigation feedback.

[0032] The intracardiac electrodes (e.g. Figure 1 The digitized intramyocardial electrogram signal 202 obtained by the exposed conductor 112 is input into the signal processor 204. The signal processor 204 can apply various digital filters, such as those described above with respect to Figure 1 The signal processor 204 may further normalize the digitized intramyocardial electrogram signal 202, segment the intramyocardial electrogram signal 202, and / or apply other transformations.

[0033] The signal processor 204 outputs processed intramyocardial electrogram data 206, which is input into a feature extractor 208. The feature extractor 208 extracts signal features of the intramyocardial electrogram data, such as R peaks and ST segment maximums, and outputs the intramyocardial electrogram data labeled with the extracted features 210. The feature extractor 208 may index the extracted features with a timestamp and / or a sample number. The extracted features 210 may further include features describing the quality of the signal and feature extraction, such as signal peaks, noise levels, and SNR. Additional details regarding the extracted features 210 are described above with respect to Figure 1 is described and is described below with respect to Figure 5-8 Describe in detail.

[0034] The extracted features 210 are applied to the processed intramyocardial electrogram data 206 to generate electrogram data 211 with the extracted features, and the electrogram data 211 with the extracted features is input into the trained intramyocardial electrogram classifier 212. The trained intramyocardial electrogram classifier 212 can be a machine learning or deep learning based classifier or a logic based classifier, which is trained with the ground truth labeled intramyocardial electrogram and its extracted features, as described below with respect to Figure 3 Described. In some instances, the trained intramyocardial electrogram classifier 212 includes a k-nearest neighbor classification model. Additionally or alternatively, the trained intramyocardial electrogram classifier 212 may include a support vector machine model using a continuous wavelet transform. In still other instances, the trained intramyocardial electrogram classifier 212 may include other machine learning-based or deep learning-based or logic-based classification algorithms, including deep learning algorithms and neural networks, which are trained using predefined waveform characteristics in electrograms obtained at different myocardial depths.

[0035] The trained intramyocardial electrogram classifier 212 analyzes the input electrogram data 211 with the extracted features and outputs radial depth categories 214. The radial depth categories 214 can correspond to the relative depths of the electrodes. The relative depths can be reported according to a predetermined number of depth categories (e.g., radial depth positions). For example, the radial depth can be classified among five depth categories, the five depth categories including intra-atrial, sub-endocardial, mid-myocardial, sub-epicardial, and extra-epicardial (e.g., intra-epicardial). As another example, the radial depth can be classified among three depth categories, wherein the first category includes mid-myocardial, the second depth category includes both sub-epicardial and sub-endocardial, and the third depth category includes both extra-epicardial and intra-atrial. As an example, a user can choose whether to classify the output radial depth categories 214 according to five depth categories or three depth categories.

[0036] The outputted radial depth category 214 may be used to generate and output a real-time depth notification 216 (also referred to as a depth indication). The real-time depth notification 216 may include one or both of a location-based depth indicator 218 and a category-based depth indicator 220. The location-based depth indicator 218 may indicate the relative location of the electrode in the myocardium, such as "mid-myocardial". The category-based depth indicator 220 may indicate the desirability of the radial depth category 214, such as according to a visual color scheme or through a text-based message. As an example, it may be most desirable to navigate a guidewire deep within the myocardium during a MIRTH procedure, making a mid-myocardial position the most desirable, and extramyocardial depth positions (e.g., extra-epicardial and intra-atrial positions) undesirable. Thus, a mid-myocardial radial depth position may be indicated by a green notification or a notification based on the text "desirable radial position". As another example, sub-epicardial and sub-endocardial depth positions may be indicated by a yellow notification or a notification based on the text "less desirable radial position". As yet another example, extra-epicardial and intra-atrial depth positions may be indicated by a red notification or a text-based notification of "Undesirable Radial Position." Outputting a category-based depth indicator 220 may provide quick, uncomplicated feedback to the operator performing the device insertion, while outputting a location-based depth indicator 218 may provide additional information regarding device steering to remain within the myocardium.

[0037] Next, Figure 3 It shows that it can be used for training Figure 2 An example of a training summary 300 of a trained intramyocardial electrogram classifier 212 is shown. Figure 3 Previously in Figure 2 Components introduced in FIG. 3 are numbered the same and will not be reintroduced. Further, it will be appreciated that the training overview 300 is provided by way of example, and other training schemes may be used without departing from the scope of the present disclosure.

[0038] The training electrogram data 302 with extracted features can be generated for the electrogram data 211 with extracted features as described above. The training electrogram data 302 with extracted features can include more than 20,000 individual electrogram waveforms, for example, electrogram waveforms obtained using monopolar electrodes, bipolar electrodes, multipolar electrodes, or a combination thereof. The training electrogram data 302 with extracted features receives ground truth labels 304 for generating labeled electrogram data 306 for the radial depth position acquired for each waveform. The labeled electrogram data 306 is divided into a training data set 308 and a validation data set 310. For example, the training data set 308 can include a first portion of the labeled electrogram data 306, and the validation data set 310 can include a second remaining portion of the labeled electrogram data 306. The first portion can include more, less, or the same amount of labeled electrogram data 306 as compared to the amount of labeled electrogram data of the second portion. The training data set 308 is input into an untrained intramyocardial electrogram classifier 312. Based on the ground truth labels of the labeled electrogram data 306 in the training data set 308 and the extracted waveform features, the untrained intramyocardial electrogram classifier 312 learns which features are statistically associated with each radial depth position and becomes the trained intramyocardial electrogram classifier 212. The trained intramyocardial electrogram classifier 212 receives the validation data set 310 and outputs the radial depth category 214 for each waveform of the validation data set 310. The accuracy of the trained intramyocardial electrogram classifier 212 can be evaluated based on the output radial depth category 214 relative to the ground truth labels 304 of the validation data set 310.

[0039] In some instances, the training electrogram data can be generated by myocardial insertion electrode experiments. Figure 4 , shows an electrogram representation overview 400 of monopolar electrode data from a myocardial insertion experiment relative to radial depth. The electrogram representation overview 400 includes a depth measurement 402, a surface ECG 404, an intracardiac electrogram 406, a relative depth indicator 418, and an axial depth diagram 420. The axial depth diagram 420 shows a short-axis view of the ventricle, including Figure 1 The myocardium 104, epicardium 103 and endocardium 105 introduced in Figure 1As explained, the myocardium extends between the epicardium 103 and the endocardium 105. The axial depth diagram 420 further illustrates relative depth indicators 418 in each relative position relative to the myocardium 104, the epicardium 103, and the endocardium 105. The relative depth indicators 418 include an extra-epicardial region 408 (represented by a light shaded fill pattern), a sub-epicardial region 410 (represented by a diagonal shaded fill pattern), a mid-myocardial region 412 (represented by a dark shaded fill pattern), a sub-endocardial region 414 (represented by a vertical shaded fill pattern), and an intracavitary region 416 (represented by an intermediate shaded fill pattern).

[0040] With the electrodes (e.g. Figure 1 As an exposed conductor 112) moves from an epicardial surface (e.g., epicardium 103) through the myocardium, the intracardiac electrogram 406 changes with each minimal change in depth from the epicardial surface. The intracardiac electrogram 406 has extractable features that can be classified based on relative depth. For example, as will be described in detail herein, an intracardiac electrogram 406 obtained in a subepicardial region 410 has a dominant R wave. As another example, an intracardiac electrogram 406 obtained in a mid-myocardial region 412 has a maximum elevation in the ST segment. Thus, an intramyocardial electrogram classifier (e.g., Figure 3 untrained intramyocardial electrogram classifier).

[0041] Now it will be relative to Figure 5-8 Describe example intramyocardial electrogram features that can be extracted (e.g., by Figure 2 The feature extractor 208 of the EDEN algorithm 200). First, go to Figure 5 , shows a first graph 500 showing a first image of a cavity at a depth within the cavity through a monopolar electrode (e.g., Figure 11 and 1 . The first graph 500 shows a plurality of waveforms of an intramyocardial electrogram signal 502 measured from an exposed conductor 112 of the first graph 500. The vertical axis of the first graph 500 represents the signal amplitude in voltage (e.g., V), while the horizontal axis represents the time in seconds (e.g., s), as marked. The second graph 504 shows a single waveform of the intramyocardial electrogram signal 502 corresponding to the portion of the first graph 500 within the box 506. The vertical axis of the second graph 504 represents the signal amplitude in voltage, while the horizontal axis represents the time in milliseconds (e.g., ms), as marked. The second figure 504 further shows multiple characteristic indicators and time points of interest, including: TP interval 508, which includes a duration between 0 milliseconds and time t1; P wave 510, which occurs between time t1 and time t2; PQ interval 512, which includes a duration between time t2 and time t3; QRS complex 514, which occurs between time t3 and time t4; ST segment 516, which includes a duration between time t4 and time t5; and T wave 520, which occurs between time t5 and time t6.

[0042] Further, T wave amplitude 522, P wave amplitude 524, and S wave amplitude 526 are indicated by dashed lines on the second graph 504. At the intracavitary depth, T wave amplitude 522 is the maximum signal amplitude of the intramyocardial electrogram signal 502, and S wave amplitude 526 is the minimum signal amplitude of the intramyocardial electrogram signal 502, as shown in signal amplitude range 530. Classifier (e.g., Figure 2 The trained intramyocardial electrogram classifier 212 can learn these characteristic signal amplitude features as well as the shapes of the various waves and the durations of the various intervals described above to identify electrograms obtained at intracavitary depth during guidewire insertion in real time.

[0043] Continue to Figure 6 , shows a first graph 600 showing a first image of a myocardial infarction at mid-myocardial depth via a monopolar electrode (eg, Figure 1604 shows a single waveform of the intramyocardial electrogram signal 602 corresponding to the portion of the first graph 600 within the box 606. The vertical axis of the second graph 604 represents the signal amplitude in voltage units, while the horizontal axis represents the time in milliseconds (e.g., ms), as marked. The second figure 604 further shows multiple characteristic indicators and time points of interest, including: TP interval 608, which includes a duration between 0 milliseconds and time t1; P wave 610, which occurs between time t1 and time t2; PQ interval 612, which includes a duration between time t2 and time t3; R peak ST segment fusion 614, which occurs between time t3 and time t4; ST segment 616, which includes a duration between time t4 and time t5; and T wave 620, which occurs between time t5 and time t6.

[0044] Further, ST segment amplitude 622, PQ interval amplitude 624, P wave amplitude 626, and TP interval amplitude 628 are indicated by dashed lines. ST segment amplitude 622 is the maximum signal amplitude of the intramyocardial electrogram signal 602, and TP interval amplitude 628 is the minimum signal amplitude of the intramyocardial electrogram signal 602, as shown in signal amplitude range 630. Classifier (e.g., Figure 2 The trained intramyocardial electrogram classifier 212 can learn these characteristic signal amplitude features as well as the shapes of the various waves and the durations of the various intervals described above to identify electrograms obtained at mid-myocardial depths during guidewire insertion in real time. For example, the classifier can use a combination of indistinguishable or R-wave-dominated QRS complex-ST segment fusion and maximally elevated ST segments to identify mid-myocardial electrograms.

[0045] Now continue to Figure 7 , shows a first graph 700 showing a first electrode (eg, Figure 11 and 1 . The first graph 700 shows a plurality of waveforms of an intramyocardial electrogram signal 702 measured from an exposed conductor 112 of the first graph 700. The vertical axis of the first graph 700 represents the signal amplitude in voltage (e.g., V), while the horizontal axis represents the time in seconds (e.g., s), as marked. The second graph 704 shows a single waveform of the intramyocardial electrogram signal 702 corresponding to the portion of the first graph 700 within the box 706. The vertical axis of the second graph 704 represents the signal amplitude in voltage, while the horizontal axis represents the time in milliseconds (e.g., ms), as marked. The second graph 704 further illustrates a plurality of characteristic indicators and time points of interest, including: a TP interval 708 comprising a duration between 0 milliseconds and time t1; a P wave 710, comprising a duration between time t1 and time t2; a PQ interval 712, comprising a duration between time t2 and time t3; an R wave ST segment deviation 714, comprising a duration between time t3 and time t4; an ST segment 716, comprising a duration between time t4 and time t5; and a T wave 720, comprising a duration between time t5 and time t6.

[0046] Further, ST segment amplitude 722, PQ interval amplitude 724, and TP interval amplitude 728 are indicated by dashed lines. ST segment amplitude 722 is the maximum signal amplitude of the intramyocardial electrogram signal 702, and TP interval amplitude 728 is the minimum signal amplitude of the intramyocardial electrogram signal 702, as shown in signal amplitude range 730. Classifier (e.g., Figure 2 The trained intramyocardial electrogram classifier 212 can learn these characteristic signal amplitude features as well as the shapes of the various waves and the durations of the various intervals described above to identify electrograms obtained at a subepicardial depth during guidewire insertion in real time. For example, the classifier can use a combination of R wave ST segment deviation and elevated ST segment to identify a subepicardial electrogram. As another example, the classifier can distinguish a subepicardial electrogram from a mid-myocardial electrogram (such as a septal electrogram) based in part on a lower ST segment amplitude 722 and a smaller signal amplitude range 730. Figure 6 Electrophysiology diagram shown).

[0047] Continue to Figure 8 , shows a first graph 800 showing a first electrode (eg, Figure 1800 ) are shown in Figure 806 . The vertical axis of the first graph 800 represents the signal amplitude in voltage (e.g., V), while the horizontal axis represents the time in seconds (e.g., s), as marked. The second graph 804 shows a single waveform of the intramyocardial electrogram signal 802 corresponding to the portion of the first graph 800 within the box 806. The vertical axis of the second graph 804 represents the signal amplitude in voltage, while the horizontal axis represents the time in milliseconds (e.g., ms), as marked. The second graph 804 further illustrates a plurality of characteristic indicators and time points of interest, including: a TP interval 808 comprising a duration between 0 milliseconds and time t1; a P wave 810 comprising a duration between time t1 and time t2; a PQ interval 812 comprising a duration between time t2 and time t3; a Q wave ST segment deviation 814 comprising a duration between time t3 and time t4; an ST segment 816 comprising a duration between time t4 and time t5; and a T wave 820 comprising a duration between time t5 and time t6.

[0048] Further, ST segment amplitude 822, PQ interval amplitude 824, and TP interval amplitude 828 are indicated by dashed lines. ST segment amplitude 822 is the maximum signal amplitude of the intramyocardial electrogram signal 802, and TP interval amplitude 828 is the minimum signal amplitude of the intramyocardial electrogram signal 802, as shown in signal amplitude range 830. Classifier (e.g., Figure 2 The trained intramyocardial electrogram classifier 212 can learn these characteristic signal amplitude features as well as the shapes of the various waves and the durations of the various intervals described above to identify electrograms obtained at a subendocardial depth during intracardiac device insertion in real time. For example, the classifier can use a combination of a Q-wave-dominated ST segment deviation and an elevated ST segment to identify a subendocardial electrogram. As another example, the classifier can distinguish a subendocardial electrogram from a subepicardial electrogram (e.g., a subepicardial electrogram) based in part on the Q wave deviation 814 of the subendocardial electrogram. Figure 7 As yet another example, the classifier may distinguish the subendocardial electrogram from the midmyocardial electrogram (e.g., Figure 6 Electrophysiology diagram shown).

[0049] The above exemplary qualitative features (eg, extracted features) and other features of intramyocardial electrograms obtained at different radial depth categories are summarized in Fig. 9900 shown in FIG. 1 . The first column 902 defines radial positions, the second column 904 defines TP segment characteristics, the third column 906 defines QRS complex characteristics, and the fourth column 908 defines ST segment characteristics for the corresponding radial position given in each row. The first row 910 summarizes the characteristics of the intramyocardial electrogram obtained at the intracavitary depth position, the second row 912 summarizes the characteristics of the intramyocardial electrogram obtained at the subendocardial depth position, the third row 914 summarizes the characteristics of the intramyocardial electrogram obtained at the mid-myocardial depth position, the fourth row 916 summarizes the characteristics of the intramyocardial electrogram obtained at the subepicardial depth position, and the fifth row 918 summarizes the characteristics of the intramyocardial electrogram obtained at the extraepicardial depth position.

[0050] As shown in the first row 910, the intramyocardial electrogram obtained at the intracavitary deep position includes an isoelectric TP segment (second column 904), a QS wave (third column 906), and an isoelectric ST segment (fourth column 908). As shown in the second row 912, the intramyocardial electrogram obtained at the subendocardial deep position includes a slight depression of the TP segment (second column 904), a dominant Q wave (third column 906), and a slight elevation of the ST segment (fourth column 908). As shown in the third row 914, the intramyocardial electrogram obtained at the mid-myocardial deep position includes the maximum depression of the TP segment (second column 904), a dominant R wave or an indistinguishable QRS complex (third column 906), and a maximum elevation of the ST segment (fourth column 908). As shown in the fourth row 916, the intramyocardial electrogram obtained at the subepicardial deep position includes a slight depression of the TP segment (second column 904), a dominant R wave (third column 906), and a slight elevation of the ST segment (fourth column 908). As shown in the first row 910, the intramyocardial electrogram obtained at the extra-epicardial depth position includes an isoelectric TP segment (second column 904), a dominant R wave (third column 906), and an isoelectric ST segment (fourth column 908).

[0051] While the endocardial (e.g., first row 910) and extraepicardial (e.g., fifth row 918) qualitative features are the same for both the TP segment (e.g., second column 904) and the ST segment (e.g., fourth column 908), they are different for the QRS complex (e.g., third column 906). As another example, the subendocardial (e.g., second row 912) and subepicardial (e.g., fourth row 916) qualitative features are also different for the individual QRS complexes (e.g., third column 906). Thus, at least one feature is different for all five radial depth categories, which enables a trained machine learning or deep learning based classifier (e.g., Figure 2 The trained intramyocardial electrogram classifier 212) is capable of identifying the radial depth category of electrograms obtained at unknown radial depths.

[0052] Further, using right ventricular pacing modeling, the characteristic features summarized in Table 900 can be present even in the presence of certain conduction abnormalities (such as left bundle branch block, which is characteristic of cardiomyopathy). Therefore, the radial depth category of the intracardiac electrogram can be effectively determined by the systems and methods described herein even in a diseased heart. In addition, the QR features listed in Table 900 can depend on the direction of the depolarization wavefront through the tissue and whether the site has rapid endocardial conduction (producing endocardial Q). Some features may also depend on the sampling rate and the time scale drawn, such as indistinguishable RS waves of the mid-myocardial electrogram. A machine learning or deep learning based classifier can be trained to recognize such changes and provide robust identification of radial depth categories.

[0053] Fig.10 A group of graphs 1000 is shown, which shows time-based changes in intracardiac electrograms recorded by a mid-myocardial electrode maintained in a non-moving position and individual electrogram waveforms 1002 obtained over time. The individual electrogram waveforms 1002 include a first electrogram waveform 1004 measured by the mid-myocardial electrode at a first earliest time (e.g., 0 minutes), a second electrogram waveform 1006 measured by the mid-myocardial electrode at a second time later than the first time (e.g., 2.5 minutes), and a third electrogram waveform 1008 measured by the mid-myocardial electrode at a third latest time (e.g., 9.5 minutes). As can be seen by comparing the first electrogram waveform 1004, the second electrogram waveform 1006, and the third electrogram waveform 1008, the amplitude of the ST segment decreases over time, while the amplitudes of the TP and PQ segments increase over time. Further, the S wave develops and deepens over time.

[0054] Characteristic time-based amplitude changes (such as those discussed above) can be tracked over time for different ECG features to generate a graph set 1000. Graph set 1000 includes a TP interval graph 1010, a PQ interval graph 1012, an ST interval graph 1014, a P peak graph 1016, a Q peak graph 1018, and a T peak graph 1020. The vertical axis of each graph set 1000 represents amplitude (e.g., in millivolts, mV), while the horizontal axis represents time (e.g., in minutes per protein). The TP interval graph 1010 shows that the TP interval amplitude increases over time, especially within approximately the first 15 minutes of the electrogram recorded at the same mid-myocardial location. The PQ interval graph 1012 and the P peak graph 1016 show similar trends. The Q peak graph shows that the amplitude of the Q peak increases slightly within approximately the first 10 minutes, and then gradually decreases over the next approximately 20 minutes. The ST interval graph 1014 shows an approximately exponential decline with a half-life of approximately 12 minutes.

[0055] The graph set 1000 may be stored in a memory, such as as a component of a deep navigation algorithm (e.g., Figure 1 Depth Navigation Algorithm 128 and / or Figure 2 The EDEN algorithm 200 may be configured to include a portion of the EDEN algorithm 200 such that the processor may consider time-based changes in the intramyocardial electrogram during intracardiac device insertion, such as in response to the intramyocardial electrode not moving. For example, movement (or lack thereof) may be based on a sensor (e.g., an accelerometer) in a guidewire torquer (e.g., Figure 1 The intramyocardial electrogram is determined by the output of the sensor 146 of the intramyocardial electrogram. It is understood that the intramyocardial electrogram is a local phenomenon, and small (e.g., less than 1 mm) position changes of the intramyocardial electrode can avoid time-based changes that result in constant positioning of the intramyocardial electrode. Therefore, sub-millimeter manipulation of the intramyocardial electrode can restore the characteristic "time zero" feature of the recorded intramyocardial electrogram.

[0056] Fig.11 A flow chart of an example method 1100 for providing depth navigation feedback during insertion of an intracardiac device (eg, a guidewire or ablation catheter) into a heart is shown. The method 1100 is related to Figure 1 The method 1100 is described with reference to the components of the EDEN system of FIG. 1100 , but it is understood that the method 1100 may be implemented using other systems and components without departing from the scope of the present disclosure. The method 1100 may be performed by a processor according to instructions stored in a non-transitory memory of the EDEN system (e.g., Figure 1 The method 1100 may include executing a depth navigation algorithm such as Figure 1 Depth Navigation Algorithm 128 and / or Figure 2 EDEN algorithm 200.

[0057] At 1102, method 1100 includes acquiring an intramyocardial electrogram (e.g., an intracardiac electrogram signal) via one or more electrodes of an intracardiac device during insertion into the myocardium (e.g., during insertion into a heart muscle or chamber). For example, the one or more electrodes may be Figure 1 The exposed conductor 112 of the guidewire 110 can be substantially continuously acquired electrograms. Figure 1 As described, the signal processor (e.g., Figure 1The analog signal from the electrode is filtered, amplified and converted into a digital signal by a signal processor 122 that is electrically coupled to the electrode and electronically communicates with or is integrated with the computing device. Alternatively, the electrodes may be photoelectrically coupled. The computing device may receive a continuous real-time digital signal from the signal processor, or receive a sliding window segment of the digital signal substantially in real time. For example, each sliding window segment may contain 0.2 seconds of digital signal. The digital signal may be further normalized, transformed and serially connected into an electrogram segment containing a complete waveform (e.g., a heartbeat cycle).

[0058] At 1104, method 1100 includes determining whether the signal quality is greater than a quality threshold. The quality threshold can be a pre-programmed value or modifier stored in a non-transitory memory that distinguishes low-quality signals that may interfere with depth navigation feedback or at least do not provide recognizable input from high-quality signals that may include characteristic waveform features. As an example, the quality threshold can be a threshold SNR. As another example, the quality threshold can be a threshold noise level (e.g., a threshold noise floor value). Thus, the processor can determine the signal quality (e.g., SNR and / or noise level) of each electrogram segment of the digital signal.

[0059] In the event that the signal quality is not greater than the quality threshold, method 1100 proceeds to 1106 and includes not processing the low quality signal. For example, the portion of the digital signal that does not exceed the quality threshold can be removed from further processing and evaluation. Then, method 1100 can return so that additional intramyocardial electrograms can be acquired.

[0060] In the event that the signal quality is greater than the quality threshold, method 1100 proceeds to 1108 and includes performing a feature extractor (e.g., Figure 2 The feature extractor 208 of the present invention extracts electrogram features from each electrogram segment in real time. For example, the electrogram can be segmented, and the feature extractor can extract signal features of each electrogram segment, such as peak and segment maximum / minimum values, interval duration, etc., and output electrogram data labeled with the extracted features. The extracted features are relative to the Figure 5-8 Additionally or alternatively, the feature extractor may index the extracted features with the timestamp and / or sample number of the corresponding electrogram segment.

[0061] At 1110, method 1100 includes inputting each electrogram segment having the extracted electrogram features into a classification algorithm. Figure 3 As described, electrograms and their extracted features can be used for classification algorithms (e.g., Figure 2 and 3The trained intramyocardial electrogram classifier 212 is trained by an experienced cardiologist according to the radial depth categories collected from these electrograms. Figure 2 As described, in some examples, the classification algorithm may use a k-nearest neighbor classification model and / or a support vector machine to analyze the input electrogram segments with the extracted electrogram features and output radial depth categories obtained by the electrogram segments. For example, the classification algorithm may compare the input electrogram segments with the extracted electrogram features with the characteristic features corresponding to each radial depth category (such as Fig. 9 and the qualitative characteristics described above).

[0062] In some examples, the method 1100 can further include identifying movement of the intracardiac device and thereby identifying movement or lack of movement of the electrode. For example, the persistence of the movement of the electrode can be monitored, and the quality of the radial depth indication can be indicated based on the persistence of the monitored movement. Fig.10 As described, when an electrode remains stationary for a long period of time (e.g., several minutes), the recorded electrogram undergoes predictable time-dependent changes. Therefore, a deep navigation algorithm can use the characteristic time-dependent changes stored in memory and apply various transformations and / or compensations based on the amount of time that the electrode does not move. Movement (or lack of movement) of the electrode can be determined based on noise in the electrogram signal and / or measured changes in guidewire torque (e.g., determined based on feedback from a sensor such as an accelerometer).

[0063] In some instances, the depth navigation algorithm can evaluate sequential waveforms (e.g., sequential electrograms) to distinguish radial depth categories with similar characteristic features, such as subepicardial and subendocardial or intra-atrial and extraepicardial. For example, comparing sequential waveforms can indicate which direction the sensing electrode is coming from. As an illustrative example, it is expected that the subepicardial electrogram is measured after the extraepicardial electrogram, rather than the subendocardial electrogram. In some instances, an indication of the direction of travel of the electrode (and therefore the intracardiac device) can be output as part of the real-time radial depth indication explained below, wherein the direction of travel of the intracardiac device is determined by evaluating the sequential electrogram. In still other instances, the sequential waveform / electrogram of the intracardiac electrogram signal can be evaluated together with the geometric position of the electrode to determine the cumulative trajectory path of the electrode (e.g., the trajectory / path of the electrode changing over time). The geometric position of the electrode can be determined based on an image (such as an X-ray image) of the electrode / intracardiac device obtained by an imaging mode. In some instances, an indication of the cumulative trajectory path of the electrode (and therefore the intracardiac device) can be output as part of the real-time radial depth indication explained below.

[0064] At 1112, method 1100 includes receiving a radial depth category output from a classification algorithm. As an example, the radial depth category can be one of intra-atrial, sub-endocardial, mid-myocardial, sub-epicardial, and extra-epicardial, such as when the classification algorithm is trained to distinguish five different radial depth categories. As another example, the radial depth category can be one of mid-myocardial, sub-epicardial / sub-endocardial (e.g., within the vicinity of the border of the myocardium), and extra-epicardial / intra-atrial (e.g., outside the myocardium), such as when the classification algorithm is trained to distinguish three different radial depth categories.

[0065] At 1114, method 1100 includes outputting a real-time radial depth indication of the radial depth of the intracardiac device based on the radial depth category output, also referred to as a real-time notification. For example, the real-time notification can be output to a display device (e.g., Figure 1 In addition or alternatively, the real-time notification may be an audio notification output through a speaker system and / or tactile feedback output through a touch-based system. For example, the visual notification may include a text-based message, a graphic, or other visual indicator. The audio notification may include a verbal message, a tone, or other audio indicator. The tactile feedback may include vibration or other tactile feedback.

[0066] In some examples, outputting a real-time notification of the device radial depth based on a radial depth category output includes outputting a location-based notification, as optionally indicated at 1116. For example, the location-based notification may describe the radial depth category (e.g., the radial depth location of the intracardiac device / electrode relative to the epicardium, myocardium, and endocardium). Additionally or alternatively, outputting a real-time notification of the device radial depth based on a radial depth category output includes outputting a category-based notification, as optionally indicated at 1118. The category-based notification may indicate the desirability of the device radial depth location, such as a mid-myocardial radial depth category of "desirable radial depth location," a subendocardial and subepicardial radial depth category of "less desirable radial depth location," and an extra-epicardial and intra-atrial radial depth category of "undesirable radial depth location." Additionally or alternatively, the category-based notification may include a graphic-based notification, such as different colors or patterns for indicating different radial depth categories. In still further examples, the radial depth indication / notification may include an indication of the direction of travel of the electrode, the cumulative trajectory path of the electrode, and / or an indication that the electrode is currently stationary.

[0067] Method 1100 can then loop back so that real-time notification of the device radial depth can be updated as the device position changes as new intramyocardial electrograms are acquired.

[0068] In this way, the electrocardiogram radial depth navigation system can provide real-time feedback about the radial position of an intracardiac device (such as a guidewire) during cardiac surgery, which fills the gap in position information provided by imaging techniques such as fluoroscopy and echocardiography. Therefore, the accuracy and efficiency of the procedure can be improved, while the difficulty of performing the procedure can be reduced. Further, by using a deep navigation algorithm based on machine learning or deep learning or logic, the efficiency of the training process can be improved. Still further, the deep navigation algorithm can also provide efficient electrogram processing so that radial depth navigation feedback is provided substantially in real time. For example, the electrogram features of the electrogram acquired by the electrode of the intracardiac device can be extracted in real time when the electrocardiogram is acquired, and the radial depth position of the electrode can be classified in real time based on the extracted electrogram features when the electrogram is acquired. A notification of the radial depth position can be output in real time when the electrogram is acquired. In other words, the first electrogram (or the first series of electrograms) can be processed when the second electrogram (or the second series of electrograms) is acquired to extract the electrogram features, classify the radial depth position, and output a notification of the radial depth position.

[0069] The technical effect of providing real-time radial position feedback during the insertion of an intracardiac device into the myocardium of the heart is that a target myocardial position can be achieved more easily and accurately by a catheter-based cardiac implant or other intracardiac device. Another technical effect of outputting a radial depth indication relative to the myocardium of the heart based on electronic processing of intracardiac electrogram signals acquired by electrodes of the intracardiac device is that the processing efficiency of a computing device (e.g., a computing device 120 executing a depth navigation algorithm) may be improved due to reduced reliance on imaging techniques such as fluoroscopy or echocardiography. For example, when providing a radial depth indication as described herein (e.g., based on electronic processing of intracardiac electrogram signals), since feedback on the position of the intracardiac device can be provided by intracardiac electrogram signals instead of (or in addition to) acquired images, fewer images can be acquired and / or the processing of acquired images can be reduced. In addition, since the accuracy of insertion can be improved, thereby preventing or reducing improper placement of the intracardiac device, the duration of the operation (e.g., inserting the intracardiac device into the heart) can be reduced. For example, the reduced duration of the operation can improve processing efficiency by reducing the number of images acquired and / or processed.

[0070] The present disclosure also provides support for a method, the method comprising: when an intracardiac device is inserted into a muscle or chamber of a heart, acquiring an intracardiac electrogram signal through an electrode of the intracardiac device, and outputting a radial depth indication relative to a myocardium of the heart based on electronic processing of the intracardiac electrogram signal. In a first instance of the method, outputting the indication based on electronic processing comprises determining a radial depth category of a position of the intracardiac device relative to the myocardium of the heart based on the intracardiac electrogram signal, the determination comprising: segmenting electrograms from the intracardiac electrogram signal, and extracting waveform features of each segmented electrogram. In a second instance of the method (optionally including the first instance), determining the radial depth category of the position of the intracardiac device relative to the myocardium of the heart based on the intracardiac electrogram signal is based on a classifier, the classifier being trained to generate an output including the radial depth category in response to the extracted waveform features. In a third instance of the method (optionally including one or both of the first and second instances), the classifier comprises a machine learning based classifier. In a fourth instance of the method (optionally including one or more or each of the first to third instances), the classifier includes a k-nearest neighbor classification model. In a fifth instance of the method (optionally including one or more or each of the first to fourth instances), the classifier includes a deep learning-based classifier or a logic-based classifier. In a sixth instance of the method (optionally including one or more or each of the first to fifth instances), the classifier is trained with radial depth-labeled electrograms and waveform features extracted therefrom. In a seventh instance of the method (optionally including one or more or each of the first to sixth instances), determining the radial depth category of the position of the intracardiac device relative to the myocardium of the heart based on the intracardiac electrogram signal includes classifying the radial depth category of the intracardiac device into one of the following: mid-myocardium, subepicardium, subendocardium, intra-atrial, and extra-epicardium. In an eighth instance of the method (optionally including one or more or each of the first to seventh instances), determining the radial depth category of the position of the intracardiac device relative to the myocardium of the heart based on the intracardiac electrogram signal includes classifying the radial depth category of the guidewire into one of: a first depth category, the first depth category including a mid-myocardial position; a second depth category, the second depth category including a subepicardial position and a subendocardial position; and a third depth category, the third depth category including an intra-atrial position and an extra-epicardial position.In a ninth example of the method (optionally including one or more or each of the first to eighth examples), outputting the radial depth indication based on the radial depth category includes outputting at least one of a location-based depth indicator and a category-based depth indicator. In a tenth example of the method (optionally including one or more or each of the first to ninth examples), the category-based depth indicator indicates the desirability of the radial depth category for the location of the intracardiac device. In an eleventh example of the method (optionally including one or more or each of the first to tenth examples), the method further includes: monitoring the continuity of movement of the electrode, and providing the quality of the radial depth indication based on the continuity of the monitored movement. In a twelfth example of the method (optionally including one or more or each of the first to eleventh examples), the method further includes: evaluating a sequential electrogram of the intracardiac electrogram signal to determine a direction of travel of the intracardiac device. In a thirteenth instance of the method (optionally including one or more or each of the first to twelfth instances), the method further comprises: filtering the intracardiac electrogram signal. In a fourteenth instance of the method (optionally including one or more or each of the first to thirteenth instances), the radial depth indication comprises an indication of a relative position in a single dimension between an endocardial surface and an epicardial surface of the heart. In a fifteenth instance of the method (optionally including one or more or each of the first to fourteenth instances), the method further comprises evaluating a sequential electrogram of the intracardiac electrogram signal and a geometric position of the electrode to determine a cumulative trajectory path of the electrode of the intracardiac device, the geometric position being determined based on an image obtained by an imaging mode, and wherein the radial depth indication comprises an indication of the cumulative trajectory path.

[0071] The present disclosure also provides support for a system, the system comprising: an intracardiac device, the intracardiac device comprising one or more electrodes; and a computing device, the computing device storing instructions in a non-transitory memory, the instructions, when executed, causing the computing device to: extract electrogram features of the acquired electrogram in real time when acquiring an electrogram through the one or more electrodes; classify the radial depth positions of the one or more electrodes in real time based on the extracted electrogram features when acquiring the electrogram; and output a notification of the radial depth position in real time when acquiring the electrogram. In a first instance of the system, the intracardiac device comprises a guidewire or a catheter. In a second instance of the system (optionally including the first instance), the guidewire comprises an insulating region and a non-insulating region, wherein the non-insulating region comprises the one or more electrodes. In a third instance of the system (optionally including one or both of the first and second instances), when acquiring the electrogram, classifying the radial depth position of the one or more electrodes in real time based on the extracted electrogram features is based on a classifier, and the classifier is trained to generate an output including the radial depth position in response to the extracted electrogram features. In a fourth instance of the system (optionally including one or more or each of the first to third instances), the classifier includes a machine learning-based classifier, a deep learning-based classifier, or a logic-based classifier. In a fifth instance of the system (optionally including one or more or each of the first to fourth instances), the radial depth position includes a relative position in a single dimension between the endocardial and epicardial surfaces of the heart.

[0072] As used herein, an element or step described in singular form and preceded by the word "a" or "an" should be understood as not excluding a plurality of said elements or steps, unless such exclusion is explicitly stated. In addition, the present invention's reference to "one embodiment" is not intended to be interpreted as excluding the existence of another embodiment that also incorporates the described features. In addition, unless explicitly stated to the contrary, an embodiment that "includes," "comprises," or "has" an element or multiple elements with a particular characteristic may include another such element that does not have the characteristic. The terms "include" and "wherein" are used as concise equivalents of the corresponding terms "include" and "wherein." In addition, the terms "first," "second," and "third," etc., are used only as markers and are not intended to impose numerical requirements or specific positional order on their objects.

[0073] This written description uses examples to disclose the invention, including the best mode, and also to enable one of ordinary skill in the art to practice the invention, including making and using any devices or systems and performing any incorporated methods. The patentable scope of the invention is defined by the claims, and may include other examples that occur to one of ordinary skill in the art. Such other examples are intended to be within the scope of the claims if they have structural elements that do not differ from the literal language of the claims or if they include equivalent structural elements with insubstantial differences from the literal language of the claims.

Claims

1. A method comprising: During insertion of the intracardiac device into a muscle or chamber of the heart, acquiring (1102) intracardiac electrogram signals through electrodes of the intracardiac device; and An indication of radial depth relative to a myocardium of the heart is output (1114) based on electronic processing of the intracardiac electrogram signal.

2. The method of claim 1, wherein outputting the indication based on electronic processing comprises determining (1112) a radial depth category of a position of the electrode of the intracardiac device relative to the myocardium of the heart based on the intracardiac electrogram signal, the determining comprising: segmenting an electrogram from the intracardiac electrogram signal; as well as Waveform features of each segmented electrogram are extracted (1108).

3. A method according to claim 2, wherein determining the radial depth category of the position of the electrode of the intracardiac device relative to the myocardium of the heart based on the intracardiac electrogram signal is based on a classifier, which is trained to generate an output including the radial depth category in response to the extracted waveform features.

4. The method of claim 3, wherein the classifier comprises a machine learning based classifier or a logic based classifier.

5. The method according to claim 3 or 4, wherein the classifier comprises a k-nearest neighbor classification model.

6. The method according to any one of claims 3 to 5, wherein the classifier is trained using radial depth labeled electrograms and waveform features extracted therefrom.

7. The method according to any one of claims 2 to 6, wherein determining the radial depth category of the position of the electrode of the intracardiac device relative to the myocardium of the heart based on the intracardiac electrogram signal includes classifying the radial depth category of the electrode of the intracardiac device into one of the following: mid-myocardial, subepicardial, subendocardial, intra-anterior chamber, and extra-epicardial.

8. A method according to any one of claims 2 to 6, wherein determining the radial depth category of the position of the electrode of the intracardiac device relative to the myocardium of the heart based on the intracardiac electrogram signal includes classifying the radial depth category of the electrode of the intracardiac device into one of the following: a first depth category, the first depth category including a mid-myocardial position; a second depth category, the second depth category including a subepicardial position and a subendocardial position; and a third depth category, the third depth category including an intra-atrial position and an extra-epicardial position.

9. The method of any one of claims 2 to 8, wherein outputting the radial depth indication based on the radial depth category comprises outputting at least one of a position-based depth indicator (1116) and a category-based depth indicator (1118).

10. The method of claim 9, wherein the category-based depth indicator indicates the desirability of the radial depth category for the location of the electrode of the intracardiac device.

11. The method of any one of claims 1 to 10, further comprising monitoring continuity of the movement of the electrode, and providing a quality of the radial depth indication based on the monitored continuity of the movement.

12. The method according to any one of claims 1 to 11, further comprising filtering the intracardiac electrogram signal and / or evaluating a sequential electrogram of the intracardiac electrogram signal to determine a direction of travel of the electrode of the intracardiac device.

13. The method of any one of claims 1 to 12, wherein the radial depth indication comprises an indication of the relative position of the electrode in a single dimension between endocardial and epicardial surfaces of the heart.

14. The method according to any one of claims 1 to 12, further comprising evaluating a sequential electrogram of the intracardiac electrogram signal and a geometric position of the electrode to determine a cumulative trajectory path of the electrode of the intracardiac device, the geometric position being determined based on an image obtained by an imaging mode, and wherein the radial depth indication includes an indication of the cumulative trajectory path.

15. A system comprising: An intracardiac device (110) comprising one or more electrodes (112); and A computing device (120) storing instructions in a non-transitory memory (126) that, when executed, cause the computing device to: When acquiring an electrogram through the one or more electrodes, extracting (1108) electrogram features of the acquired electrogram in real time; When acquiring the electrogram, classifying the radial depth positions of the one or more electrodes in real time based on the extracted electrogram features (1112); and When acquiring the electrogram, a notification of the radial depth position is output (1114) in real time.

16. The system of claim 15, wherein the intracardiac device comprises a guidewire or a catheter, and wherein the one or more electrodes comprise a monopolar electrode or a bipolar electrode.

17. The system of claim 16, wherein the guidewire comprises an insulated region and a non-insulated region, wherein the non-insulated region comprises the one or more electrodes.

18. A system according to any one of claims 15 to 17, wherein when acquiring the electrogram, real-time classification of the radial depth position of the one or more electrodes based on the extracted electrogram features is performed based on a classifier, and the classifier is trained to generate an output including the radial depth position in response to the extracted electrogram features.

19. The system of claim 18, wherein the classifier comprises a machine learning based classifier, a deep learning based classifier, or a logic based classifier.

20. The system of any one of claims 15 to 19, wherein the radial depth position comprises a relative position in a single dimension between endocardial and epicardial surfaces of the heart.

Images