Method for analyzing electrograms
By introducing a cost function based on the characteristics of activation candidates in intracardiac electrograms, the jump length and sequence selection are optimized, the complexity of detection dominant rhythms in atrial fibrillation is solved, and the detection accuracy and accuracy of treatment decisions are improved.
Patent Information
- Application Number
- CN202410713916.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2023-06-06
- Filing Date
- 2024-06-04
- Publication Date
- 2024-12-06
AI Technical Summary
During atrial fibrillation, the complexity and confusion of detecting local and far-field activation rhythms makes it difficult for prior art to accurately identify dominant rhythms, affecting treatment decisions.
By introducing a cost function, the cost is allocated based on the characteristics and similarity of the activation candidates, the jump length and sequence selection are optimized, and the data of multiple channels are combined to optimize consistency and physical location to identify physiologically effective activation sequences.
It improves the detection accuracy of local activation of atrial fibrillation, reduces far-field potential interference, enhances the recognition ability of dominant rhythms, and supports more accurate treatment plan design.
Smart Images

Figure CN119074006A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to a method for analyzing an electrogram according to the general part of claim 1 and to a control system according to claim 14 configured to perform the proposed method. Background Art
[0002] The present method is particularly concerned with atrial fibrillation and atrial flutter. Electrically, atrial fibrillation is a chaotic activation of the myocytes of the atria. During atrial fibrillation, the atria contribute minimally to cardiac function. Therefore, atrial fibrillation reduces cardiac output but is not an imminent danger. However, when it becomes chronic, atrial fibrillation is associated with increased morbidity and mortality.
[0003] During atrial fibrillation, there may be competitive activation rhythms in the atrium. Some of these rhythms may be stable, while other rhythms appear and disappear quickly. Different methods for assessing the severity of atrial fibrillation and evaluating treatment options include detecting the rhythm in one or two atria by detecting a sequence representing the rhythm. However, the more chaotic and complex atrial fibrillation is, the more competitive rhythms may exist in a given area in the atrium. Therefore, the measurement of the electrodes placed in the atrium will include different rhythms of local cell depolarization and currents conducted by electrical conduction that is not based on cell depolarization. For example, during a 3D mapping process, detecting local activation and deriving one or more dominant local rhythms allows a better understanding of atrial fibrillation. However, it is challenging to detect which activation is local and which or which rhythms cause these activations, especially during atrial fibrillation.
[0004] The known prior art (WO 2015 / 149153 A1) on which the present invention is based relates to a method according to the general part of claim 1 .
[0005] WO 2015 / 149153 A1 proposes an annotation method in which, in a first calculation step, activation candidates are identified and an overall estimate of the average activation interval (hereinafter referred to as "jump length") is calculated. Then in a second calculation step, sequences of activation candidates are identified so that they minimize the total difference between the estimates of the individual activation intervals and the total activation interval. The prior art achieves this by using a cost function that assigns costs to jumps between activation candidates. In particular, a cost is assigned to each possible jump between activation candidates, and a path through the time with the lowest cost is identified. This overall minimization can be achieved by a computational method for the shortest path search known in computational graph theory. An example of a suitable shortest path search is the Djikstra algorithm.
[0006] Known algorithms provide a good basis for annotating the dominant rhythm in a channel of a (here intracardiac) electrogram. Nevertheless, there is still room to improve the algorithm to improve the detection of local rhythms. Summary of the invention
[0007] The present invention is based on the problem of improving the known methods so as to achieve further optimization with respect to the challenges described.
[0008] The above-mentioned object is solved by the characterizing features of claim 1 .
[0009] The main realization of the present invention is that further information can be used to decide which activation candidates form a physiologically valid sequence. This further information comes in the form of properties of the activation candidates themselves. It is proposed to convert these properties into costs for the activation candidates. Exemplarily using a cost function to find the sequence being searched, then instead of or in addition to the cost of the jumps between activation candidates, the path through time depends on the cost of each activation candidate that is part of the sequence.
[0010] Several properties have been identified that can be used for such costs. For example, the morphology of activation candidates for one rhythm may have higher similarity than the morphology of activation candidates for a different rhythm. In addition, cross-channel similarities between activation candidates, in particular their timing, can be used. Local activation propagates through channels more slowly than far-field potentials, which often appear in some or all channels essentially simultaneously. In contrast, local activation sequences that do not appear in more than one channel or appear in only a few channels may be regionally restricted activation sequences, while local activation sequences that can be found in more channels or channels that are more distantly spaced may be the dominant activation rhythm being sought. The question of which rhythm to seek may also depend on clinical use, which can also be encoded into the cost.
[0011] Therefore, it is proposed to assign a cost component to the activation candidates to identify sequences based on the characteristics of the activation candidates.
[0012] Here and preferably, the electrogram is an intracardiac electrogram. The term "intracardiac" should be understood in a broad sense and relates to measurements made from the inside of the human heart and directly adjacent thereto, for example in the pulmonary veins. Alternatively, the electrogram may have been recorded by any other invasive method. In particular, epicardial measurements are included.
[0013] In detail, it is proposed that the control system assigns cost components to activation candidates, the cost of a potential sequence comprising the cost components defining the activation candidates of the corresponding potential sequence.
[0014] In a preferred embodiment according to claim 2, a cost component is also assigned to the jumps between activation candidates. This allows the optimization of consistent jump lengths while searching for rhythmic sequences and taking into account the properties of the activation candidates at the same time.
[0015] In a preferred embodiment according to claim 3, some cost components depend on data from a second channel. In particular, the cost components assigned to activation candidates may depend on the temporal distance between these activation candidates and activation candidates from another channel. Defining the costs in this way optimizes the consistency between electrodes. Such consistency is particularly advantageous in the presence of competing activation rhythms.
[0016] According to claim 4, the cost component depending on the second channel may include time-dependent and / or time-independent cost components. A time-dependent cost component is a cost component that depends on the timing of the activation candidate, so that hypothetically changing the timestamp of the activation candidate will have an impact on the cost component all other things being equal. If only the timestamp of the activation candidate is changed, the time-independent cost component will not change.
[0017] If a mapping system is not used, the control system may not have any data about the relative positions of the electrodes and the corresponding channels. It is still possible to derive which electrodes are physically close to each other by using the ordering of electrodes and channels (claim 5). For example, a rhythm that can be tracked by channels I to III but not by channels IV and V is quite unlikely to appear again in channel VI.
[0018] Claim 6 relates to a preferred embodiment, wherein the monopolar and bipolar signals are combined to detect the rhythm by using a cost component depending on the monopolar channel as a cost component for the activation candidate in the bipolar channel. This idea is based on the fact that cell depolarization is always negative, so the downward deflection of the monopolar electrode can be advantageously included in the cost component to detect the local activation.
[0019] In order to detect far-field potentials and reduce their influence on the analysis results, the cost component can depend on data from the coronary sinus electrodes and / or the surface ECG electrodes (claim 7). The coronary sinus electrodes can be used to detect the actual beating of the heart and increase the cost component of activating candidates shortly after the heartbeat, which may have a higher chance of being interfered by far fields.
[0020] An embodiment according to claim 8 involves basing the cost component of an activation candidate on the morphology of the activation candidate. According to claim 9, in particular morphological similarities between activation candidates are included in the cost component, for example by basing the cost component of a jump between activation candidates on the morphological similarity of the two activation candidates of the jump connection.
[0021] Claim 10 relates to the possibility of static and / or dynamic cost components. The static cost component is independent of the sequence, while the dynamic cost component varies with the sequence. For example, the cost component of a jump can depend on the average jump length of other jumps in the sequence, so that the overall similarity of jump lengths in the sequence is optimized instead of the similarity of jump lengths to a predefined average jump length.
[0022] According to claims 11 and 12, channels may be analyzed iteratively and the processing of a channel may depend on previously processed channels. It is possible that for at least one channel more than one iteration may be performed.
[0023] In an embodiment according to claim 13, the analysis is performed cyclically, in particular in real time, and the control system analyzes the channels cyclically, for example, every second. Since the dominant rhythm may change, it is preferably possible that the detected sequence changes with respect to the past with a new analysis cycle. The results of the previous analysis can be included in the cost to reduce the probability of detecting two different rhythms and frequently changing the detected rhythm. Including the past analysis in the cost creates an artificial barrier for changes between detected rhythms.
[0024] Another teaching of equal importance according to claim 14 relates to a control system configured to carry out the proposed method, wherein the control system is configured to receive and / or measure an electrogram.
[0025] All explanations given regarding the proposed method are fully applicable. BRIEF DESCRIPTION OF THE DRAWINGS
[0026] Below, the embodiment of the present invention is explained with reference to the accompanying drawings. This figure shows: Figure 1 Show that the sequence is found by analyzing the cost components, Figure 2 shows that cost components for multiple channels are iteratively derived according to one embodiment, Figure 3 It is shown that the cost component which is dependent on another channel is derived in a time-dependent manner. DETAILED DESCRIPTION
[0027] The accompanying drawings show the proposed method for analyzing electrograms Figure 1 In an embodiment of the method, the electrogram Figure 1 In this case, the intracardiac electrogram is preferred. Figure 1 It is generally shown how to determine a sequence 2 based on a plurality of nodes and edges between the nodes using a cost function. Figure 2 and 3 Intracardiac electrograms are shown in a rather schematic manner. Figure 1 An exemplary portion of an intracardiac electrogram is shown. Figure 1It includes the coronary sinus channel M, which is only very schematically depicted at the top, and the purely exemplary 8 bipolar atrial channels L. Corresponding intracardiac electrograms Figure 1 may have been recorded during the mapping process. Preferably, intracardiac electrograms have been recorded during atrial fibrillation Figure 1 .
[0028] The analysis is performed by a control system, such as a general purpose computer, a mapping system or a dedicated hardware device including a processor. Therefore, the proposed method is a computer-implemented method. Therefore, any method steps described herein can be performed by a control system, and the control system can be adapted accordingly.
[0029] Electrograms have been recorded through catheters inserted into the body Figure 1 The catheter includes a plurality of electrodes. Figure 1 At least one channel K of the catheter has been recorded by an electrode, preferably between the electrodes. The channel K or channels K may be monopolar, recorded by one electrode of the catheter, or bipolar, recorded between two electrodes of the catheter.
[0030] In general, the proposed method involves detecting electrograms Figure 1 The control system can display the results of the analysis to the user, in particular the surgeon. The user can use this information for diagnostic purposes and / or to design a treatment plan.
[0031] Electrodes placed in the atria will always measure local electrical depolarization of cells and potentials that reach the electrodes through electrical conduction, also called far-field potentials or far-field interference. During normal cardiac activity in healthy atria, local activations include one rhythm that corresponds to the atrial beat. During atrial fibrillation, different locations may have different rhythms, and even one location may have competing rhythms. In addition, random activations and activations from neighboring tissues will be visible in channel K of a given electrode. In order to better understand the complex composite picture of atrial fibrillation, it is desirable to annotate a single or a small number of rhythms for a given channel K, for example for subsequent creation of a map of the electrical activation of the atria.
[0032] As part of the proposed method, the control system detects an activation candidate 3 in a first channel K_a. The terms "first" and "second" are not limiting and are only used to help understand the explanation. Any channel K can be a first channel K_a, and any other channel K can be a second channel K_b as will be explained. An activation candidate 3 is a portion of the signal of the corresponding channel K that includes a peak with a certain characteristic, or is otherwise selected as likely to represent a local depolarization. Figure 2 and 3 Intracardiac electrogram Figure 1A plurality of such activation candidates 3 are shown.
[0033] A plurality of different potential sequences 4 of activation candidates 3 may be defined along the time dimension 5 of the first channel K_a. In principle, any combination of at least three, preferably at least 5, more preferably at least 10 activation candidates 3 as a subgroup of all activation candidates 3 may form a potential sequence 4. It should be understood that the number of potential sequences 4 is greater than 1 and that a sequence 2 as a subgroup of activation candidates 3 does not include all activation candidates 3 of channel k.
[0034] The control system assigns a cost comprising one or more independent cost components to a potential sequence 4. In general, the cost of a potential sequence 4 consists of cost components of the potential sequence 4, in particular cost components of elements of the potential sequence 4. For example, a sequence 2 comprising a number of activation candidates 3 distributed over a time dimension 5 may include a cost component assigned to jumps 6 between activation candidates 3, each jump 6 being the time distance between an activation candidate 3 and the next activation candidate 3 selected from the potential sequence 4. The cost component assigned to a jump 6 may simply be the length of the jump 6. Further illustratively, the cost may be the sum of the cost components. However, it should be noted that the length of the jump 6 alone is not a useful cost function, since the sum of the costs is simply the time difference between the first and last activation candidate 3.
[0035] The control system selects a sequence 2 of dominant activations from the potential sequences 4 of the first channel K_a, for example by selecting the potential sequence 4 with the lowest cost. More generally, the control system selects a sequence 2 of dominant activations that satisfies an optimization criterion based on the cost of the potential sequences 4. In particular, it is possible to find a mathematically equivalent or similar optimization criterion for the minimum cost.
[0036] Here and preferably, the first channel K_a is a bipolar channel K and / or an atrial channel L and / or is not the coronary sinus channel M.
[0037] It is proposed that the control system assigns a cost component to the activation candidate 3. The cost of the potential sequence 4 then comprises the cost component defining the activation candidate 3 of the respective potential sequence 4. Figure 1 A general example of how a sequence 2 may be selected is shown. Figure 1 7 activation candidates 3 are shown, which are shown as nodes and some possible jumps 6 between nodes. For a better understanding of the figure, not all possible jumps 6 are shown. It may be the case that jumps 6 exceeding a predefined length and / or jumps 6 below a predefined length are picked out to reduce the number of potential sequences 4. Each activation candidate 3 has an assigned cost component "c", the index being the number of the activation candidate 3. As an optional addition, Figure 1The cost components assigned to hop 6 between nodes are also shown and are labeled "j" with indices indicating which node to which node hop 6 leads from. For example, the selected sequence 2 shown with a solid line would have a cost of the sum of the cost components of activation candidates 3 numbered 1, 2, 4, 5, and 7 and the cost components of hop 6 between nodes 1 and 2, 2 and 4, 4 and 5, and 5 and 7.
[0038] While any mathematically equivalent form of assigning cost components to nodes and still assigning cost components to activation candidates 3 may be used, for example by making the cost components of jumps 6 depend on their start and target nodes, thereby hiding the cost components of nodes in the cost components of jumps 6, it should still be understood that the cost components of activation candidates 3 depend on the respective activation candidates 3 independently of their mathematical representation and can be distinguished from the cost components that depend on jumps 6, in particular the jump length CL, which is independent of the activation candidate 3 itself. Here and preferably, the cost component of the activation candidate 3 is independent of the timing of the activation candidate 3 within the channel K and / or independent of the position of the activation candidate 3 in the sequence 2.
[0039] Preferably, the cost components of the activation candidates 3 are time-invariant within channel K. Thus, if, for example, three activation candidates 3 with different morphologies are placed at 0 ms, 100 ms and 200 ms of a hypothetical channel K, if all cost components are time-invariant, then the cost of sequence 2 is the same if it is assumed that the ordering of the activation candidates 3 is changed by swapping the positions of two activation candidates 3, all other conditions being the same. Naturally, such swapping is difficult to implement in real signals.
[0040] It should also be noted that the costs including the cost components mentioned here are not zero, nor are they infinite or prohibitively high and therefore are effectively infinite. Such infinite or zero costs may exist in addition to the costs mentioned, and different parts of the costs may be zero or infinite in different iterations of the method, but the costs mentioned here and related to this description are always by design the costs that may actually result in the selection of the corresponding element of sequence 2.
[0041] Figure 2 and 3 A preferred embodiment of an algorithm for selecting a sequence 2 from a plurality of activation candidates 3 is shown. In the following, this embodiment will be used as an example to explain further advantageous embodiments of the present invention. Figure 2In FIG. 1 , as a first step, the first channel K_a (here the top atrial channel L) is first selected and processed (top right arrow). The control system detects activation candidates 3 (marked with circles) in the first channel K_a and optionally assigns a time position to each of them based on, for example, the position of the main peak of the activation candidate 3. The detection of the activation candidate 3 can be done with known algorithms, for example based on peak detection, morphological detection or other methods. Figure 2 Continuing downward in FIG. 3 , the control system assigns cost component c to activation candidate 3 .
[0042] According to one embodiment, it is proposed that the control system assigns a cost component to the jumps 6 between the activation candidates 3. Figure 2 This embodiment with a cost component j can also be seen in . The cost of a potential sequence 4 then comprises a cost component defining jumps 6 between activation candidates 3 of the corresponding potential sequence 4. The cost component assigned to jumps 6 may depend on the difference between the length of the jumps 6 and an estimate of an average jump length CL (also called cycle length). The average jump length CL may be the expected jump length CL of an expected dominant rhythm. Known ways of estimating the average jump length CL include calculation of the dominant frequency and other methods using, for example, Fourier transforms. Figure 2 It is shown that the average jump length CL of the first channel K_a can be estimated from all atrial channels L. It is generally the case that the average jump length CL of a channel K can be estimated from the same channel K or multiple channels K or another channel K, as will be explained for the second channel K_b.
[0043] In one embodiment, the potential sequences 4 are restricted by defining a start and / or end region of the sequences 2 and considering only the sequences 2 having a first and / or last activation candidate 3 within the respective region.
[0044] With the given information, it is already possible to select sequence 2 from the potential sequences 4. It is also possible to process all channels K in this way and thus analyze the electrograms Figure 1 .
[0045] However, according to one embodiment, it is proposed that the cost of the potential sequence 4 of the first channel K_a, in particular the cost component assigned to the activation candidate 3, includes a cost component that depends on data derived from the second channel K_b. In this embodiment, unlike the example given above, Figure 2 The first channel K analyzed in is the second channel K_b, and any other channel K is the first channel K_a. Alternatively, as will be explained, the topmost channel K may be analyzed twice, in which case it is irrelevant which channel K is defined as the first channel K_a. Figure 2It is shown that after analyzing the topmost channel K (here the second channel K_b), an average jump length CL estimate can be derived from the selected sequence 2 of the second channel K_b. This estimate is then used to find the sequence 2 in the first channel K_a by using data from another channel K, here the average jump length CL estimate.
[0046] Figure 3 Another preferred embodiment is shown, in which the first channel K_a, here the channel K with number II, is analyzed by assigning to the activation candidates 3 of the first channel K_a a cost component that depends on the second channel K_b (the previously analyzed channel K with number I). Here and preferably, the cost component assigned to the activation candidates 3 of the first channel K_a depends on the activation candidates 3 of the selected sequence 2 of the previously analyzed second channel K_b. Figure 3 It is shown how the cost component of the activation candidate 3 depends on the distance in time to the activation candidate 3 of the second channel K_b. If a similar sequence 2 is found in the other channel K, the chance that sequence 2 actually represents the dominant rhythm is higher.
[0047] Here and preferably, the first channel K_a and the second channel K_b are channels K of electrodes from the same catheter and / or bipolar channels K and / or atrial channels L and / or neither the first channel K_a nor the second channel K_b is the coronary sinus channel M.
[0048] According to one embodiment, it is proposed that the cost component depending on the data derived from the second channel K_b comprises a time-independent cost component and / or a time-dependent cost component. The time-independent cost component is the same, independent of the timing of the corresponding element of the potential sequence 4 to which its cost component is assigned. The above example with the average jump length CL is time-independent.
[0049] In general, preferably, the time-independent cost component depends on the difference between the hop length CL of the first channel K_a and an estimate of the average hop length CL of the second channel K_b.
[0050] The above example of a cost component for activating candidate 3 is time-dependent, because the cost component depends on the timing of activating candidate 3 relative to activation candidates 3 of other channels K. In general, preferably, the time-dependent cost component depends on the timing of activation candidate 3 of the first channel K_a compared to the timing of activation candidate 3 of the second channel K_b.
[0051] Preferably, the second channel K_b is a neighboring channel K of the first channel K_a. The control system can use the ordering of the electrodes to derive which channels K are neighboring channels K. Typically, the channel K or electrode number is selected based on the physical location of the electrode on the catheter. Therefore, a simple way to estimate the physical neighbors of the electrode is to use the provided channel K order.
[0052] Another way to assign a cost component to a potential sequence 4 is based on a unipolar signal of a bipolar channel K. This may be the case, where the first channel K_a is a bipolar channel K, and the cost of the potential sequence 4 of the first channel K_a, in particular the cost component assigned to the activation candidate 3, comprises a cost component that depends on data derived from one or both unipolar channels K of the electrodes of the bipolar channel K, in particular data on the polarization of one or more electrodes. This embodiment is based on the recognition that cellular depolarization is always negative, and therefore a downward deflection of an electrode indicates whether the activation is localized.
[0053] According to another embodiment, the cost of the potential sequence 4 of the first channel K_a, in particular the cost component assigned to the activation candidates 3, is proposed, including a cost component that depends on data derived from the coronary sinus channel M and / or the surface ECG electrodes. The coronary sinus electrodes can be used to estimate which activation candidates 3 are more likely to be far-field interferences. The cost component that depends on data derived from the coronary sinus channel M and / or the surface ECG electrodes can therefore depend on data about the time distance between the activation candidates 3 in the first channel K_a and the ventricular beats detected in the coronary sinus channel M and / or the surface ECG electrodes. As an alternative to increasing their cost component, activation candidates 3 that are highly likely to be far-field interferences can be completely removed so that they cannot be selected as part of the sequence 2.
[0054] It may also be the case that the cost component assigned to an activation candidate 3 comprises a cost component that depends on the morphology of the respective activation candidate 3. The morphology may comprise descriptive features such as amplitude and / or dV / dt slope and / or RMS value and / or frequency content and / or correlation coefficient etc., or the cost may be derived from the output of an algorithm (such as a machine learning model) by taking as at least one of its inputs the electrogram intervals.
[0055] In another embodiment, it may be the case that the clinical use of the analysis is encoded into a cost component, in particular a cost component assigned to the activation candidate 3. The clinical use may be input by the user, in particular selected from a list.
[0056] Furthermore, the cost of a potential sequence 4 may include a cost component that depends on a measure of similarity (particularly morphological similarity) between the activation candidates 3 of the first channel K_a and / or the selected sequence 2 of the previously analyzed second channel K_b. Similar activation candidates 3 may be more likely to form a valid sequence 2. Using activation candidates 3 that have already been defined as a valid sequence 2 reduces the number of bad morphologies in the comparison.
[0057] Another possibility is that the cost of the potential sequence 4 of the first channel K_a, in particular the cost components assigned to the activation candidates 3 and / or the jumps 6, comprises a static cost component independent of the potential sequence 4 and / or a dynamic cost component depending on the potential sequence 4. In many cases it is computationally easier not to use a cost component that depends on the sequence 2 itself, but, in particular for morphological comparisons, it may be advantageous to compare the morphology of the activation candidates 3 of the sequence 2 and then adapt the cost of this sequence 2.
[0058] Preferably, the dynamic cost component assigned to jumps 6 depends on a measure of similarity of jump lengths CL of corresponding potential sequences 4, thereby rewarding jumps 6 of similar length. Additionally or alternatively, the dynamic cost component assigned to activation candidates 3 may depend on a measure of similarity of activation candidates 3 of corresponding potential sequences 4.
[0059] It can be concluded that here and preferably, the analysis of channel K is not a strict forward search, but uses global optimization. Here and preferably, a graph search algorithm, such as Dijkstra, is used to select a sequence that satisfies the optimization criterion. Generally, preferably, a global optimization algorithm is used to select a sequence that satisfies the optimization criterion for the cost of a given channel.
[0060] However, for computational efficiency, it may be the case that the control system iteratively analyses channel K and that the analysis of channel K depends on the analysis of previously analyzed channel K. Preferably, the cost component of the potential sequence 4 of channel K depends on the analysis of previously analyzed channel K. More preferably, at least one channel K is sometimes analyzed twice, in particular if a measure of accuracy of the analysis has not been met, whereby the second analysis depends on the analysis of a channel K analyzed after the first analysis. This is particularly the case for the channel K that was analyzed first.
[0061] As described above, according to one embodiment, a time-dependent cost component is proposed that depends on the timing of the activation candidates 3 of the first channel K_a compared to the timing of the activation candidates 3 of the selected sequence 2 of the second channel K_b previously analyzed.
[0062] Turning to clinical use, it may be the case that a control system cyclically receives and analyzes an update channel K, the update channel K comprising a previously analyzed portion and a new portion, and that the selection of activation candidates 3 in the previously analyzed portion sometimes changes as a function of the analysis of the update channel K. This embodiment is considered highly unusual. Normally, the analysis does not change the annotations made in the previous round of analysis. However, the dominant rhythm in a given area may change over time, and this change may have announced itself in previous data without having made it into the selected sequence 2. Retrospectively re-labeling the data allows for faster reaction and better detection of such changes.
[0063] Preferably, the cost component of potential sequences 4 in the analysis of update channel K includes a cost component that depends on the selection of activation candidates 3 in the previous analyzed part. By providing a storage of previous labels in the cost component, frequent changes back and forth between two similar dominant rhythms become less likely.
[0064] Another teaching of equal importance relates to a control system configured to carry out the method according to one of the preceding claims, wherein the control system is configured to receive and / or measure electrograms Figure 1 The control system may comprise a processor and / or a memory. The control system may be a local unit with a processor, a possible user interface, etc. In one embodiment, it may also comprise a cloud processor. Thus, the control system is not necessarily limited to a single device. The control system may comprise an interface for a catheter. The control system is configured to perform the proposed method.
Claims
1. A method for analyzing an electrogram (1) via a control system, wherein the electrogram (1) has been recorded via a catheter inserted into a human body, wherein the catheter comprises a plurality of electrodes, wherein at least one channel (K) of the electrogram (1) has been recorded by the electrodes, preferably between the electrodes, wherein the control system detects activation candidates (3) in a first channel (K_a), wherein a plurality of different potential sequences (4) of activation candidates (3) can be defined along a time dimension (5) of the first channel (K_a), wherein the control system assigns costs comprising one or more independent cost components to the potential sequences (4), wherein the control system selects a dominantly activated sequence (2) from the potential sequences (4) of the first channel (K_a), wherein the control system selects the dominantly activated sequence (2) that satisfies an optimization criterion based on the cost of the potential sequences (4), Features The control system assigns cost components to the activation candidates (3), and the cost of the potential sequence (4) includes the cost components of the activation candidates (3) defining the corresponding potential sequence (4).
2. The method according to claim 1, It is characterized in that The control system assigns a cost component to a jump (6) between activation candidates (3), the cost of the potential sequence (4) comprising a cost component defining the jump (6) between activation candidates (3) of the corresponding potential sequence (4), preferably, the cost component assigned to the jump (6) depends on the difference between the length of the jump (6) and an estimate of the average jump length (CL).
3. The method according to claim 1 or 2, It is characterized in that The cost of the potential sequence (4) of the first channel (K_a), in particular the cost component assigned to the activation candidate (3), includes a cost component that depends on data derived from the second channel (K_b), preferably, the first channel (K_a) and the second channel (K_b) are channels (K) of electrodes from the same catheter and / or bipolar channels (K) and / or atrial channels (L) and / or neither the first channel (K_a) nor the second channel (K_b) is a coronary sinus channel (M).
4. The method according to claim 3, It is characterized in that The cost components that depend on the data derived from the second channel (K_b) include time-independent cost components and / or time-dependent cost components, Preferably, the time-independent cost component depends on the difference between an estimate of the jump length (CL) of the first channel (K_a) and an estimate of the average jump length (CL) of the second channel (K_b), and / or, The time-dependent cost component depends on the timing of the activation candidate (3) of the first channel (K_a) compared to the timing of the activation candidate (3) of the second channel (K_b).
5. The method according to claim 3 or 4, It is characterized in that The second channel (K_b) is a neighboring channel (K) of the first channel (K_a), and preferably, the control system uses the ordering of the electrodes to derive which channels (K) are neighboring channels (K).
6. The method according to any one of the preceding claims, It is characterized in that The first channel (K_a) is a bipolar channel (K), and the cost of the potential sequence (4) of the first channel (K_a), in particular the cost component assigned to the activation candidate (3), includes data derived from one or two monopolar channels (K) of the electrodes of the bipolar channel (K), in particular data about the polarization of one or more electrodes.
7. The method according to any one of the preceding claims, It is characterized in that The cost of the potential sequence (4) of the first channel (K_a), in particular the cost component assigned to the activation candidate (3), includes a cost component depending on data derived from the coronary sinus channel (M) and / or the surface ECG electrodes, in particular depending on data about the temporal distance between the activation candidate (3) in the first channel (K_a) and the ventricular beat detected in the coronary sinus channel (M) and / or the surface ECG electrodes.
8. The method according to any one of the preceding claims, It is characterized in that The cost components allocated to the activation candidates (3) include cost components that depend on the morphology of the corresponding activation candidates (3).
9. The method according to any one of the preceding claims, It is characterized in that The cost of the potential sequence (4) comprises a measure of similarity, in particular morphological, between the activation candidates (3) depending on the first channel (K_a).
10. The method according to any one of the preceding claims, It is characterized in that The cost of the potential sequence (4) of the first channel (K_a), in particular the cost components assigned to the activation candidates (3) and / or the jumps (6), comprises a static cost component independent of the potential sequence (4) and / or a dynamic cost component depending on the potential sequence (4), preferably, the dynamic cost component assigned to the jumps (6) depends on a measure of similarity of the jump lengths (CL) of the corresponding potential sequences (4), and / or the dynamic cost component assigned to the activation candidates (3) depends on a measure of similarity of the activation candidates (3) of the corresponding potential sequences (4).
11. The method according to any one of the preceding claims, It is characterized in that The control system iteratively analyses the channel (K), and the analysis of the channel (K) depends on the analysis of the previously analysed channel (K), preferably a cost component of a potential sequence (4) of the channel (K) depends on the analysis of the previously analysed channel (K), preferably at least one channel (K) is sometimes analysed at least twice, in particular if a measure of accuracy of the analysis has not been met, whereby the second analysis depends on the analysis of the channel (K) analysed after the first analysis.
12. The method according to claims 4 and 11, It is characterized in that The time-dependent cost component depends on the timing of the activation candidates (3) of the first channel (K_a) compared to the timing of the activation candidates (3) of the selected sequence (2) of the second channel (K_b) analyzed previously.
13. The method according to any one of the preceding claims, It is characterized in that The control system cyclically receives and analyzes an update channel (K), wherein the update channel (K) includes a previously analyzed part and a new part, wherein the selection of an activation candidate (3) in the previously analyzed part sometimes changes with the analysis of the update channel (K), and preferably, the cost component of the potential sequence (4) in the analysis of the update channel (K) includes a cost component that depends on the selection of the activation candidate (3) in the previously analyzed part.
14. A control system configured to perform the method according to one of the preceding claims, wherein the control system is configured to receive and / or measure the electrogram (1).
