Program for outputting data indicating shape of ablation lesion

Abstract

The purpose of the present invention is to provide a program for outputting data indicating the shape of an ablation lesion due to ablation in arrhythmia treatment. The present invention provides a program for outputting data indicating the shape of an ablation lesion due to ablation in arrhythmia treatment in a subject. The program is generated by a method including a step for inputting, as learning data, a plurality of pieces of data indicating an active index or a passive index of ablation in arrhythmia treatment and the shape of the ablation lesion due to ablation in arrhythmia treatment, into an artificial intelligence model, and causing the artificial intelligence model to learn.

Description

A program for outputting data showing the shape of ablation lesions

[0001] The present invention relates to a program for outputting data indicating the shape of an ablation lesion caused by ablation in arrhythmia treatment, a computer storing the program, and a computer-readable recording medium storing the program.The present invention also relates to an apparatus for outputting data indicating the shape of an ablation lesion caused by ablation in arrhythmia treatment in a subject.Furthermore, the present invention relates to a method for outputting data indicating the shape of an ablation lesion caused by ablation in arrhythmia treatment in a subject.Furthermore, the present invention relates to a method for predicting the shape of an ablation lesion caused by ablation in arrhythmia treatment.

[0002] Active indices such as output power, ablation time, and contact force have been used to assess the safety and effectiveness of radiofrequency catheter ablation in arrhythmia treatment. Combining these indices has been shown to correlate more favorably with the shape of the ablation lesion. However, these indices do not reflect external factors, such as tissue response during ablation, and therefore are insufficient for predicting the efficacy and safety of long-duration or high-power ablation. In recent years, catheters capable of directly measuring passive indices, such as local temperature and local resistance at the ablation site, have been developed. However, these techniques are inadequate because they do not reflect active indices. Furthermore, techniques have been developed to predict ablation efficacy and safety using individual active or passive indices, or variables combining multiple active indices. However, the accuracy of these techniques remains clinically insufficient. For more information, please refer to Non-Patent Documents 1 and 2 and Patent Document 1. Non-Patent Document 2 reports that combining the drop in local impedance (% Impedance-drop, a passive index) with the conventional parameters AE (ablation energy, an active index) and FTI (force-time integral, an active index) may provide a stronger correlation with lesion characteristics. Furthermore, Patent Document 1, which describes an invention related to cardiac ablation technology using machine learning and AI, describes estimating the size of the lesion formed in a patient's heart during ablation and describes the use of various data / information for this purpose.

[0003] Japanese Patent Publication No. 2022-31229

[0004] JACC Clin Electrophysiol. 2018 Jun; 4(6):707-723.J Interv Card Electrophusiol., 63, 389-398, 2022.

[0005] Therefore, an object of the present invention is to provide a program for outputting data indicating the shape of an ablation lesion caused by ablation in arrhythmia treatment; an apparatus for outputting data indicating the shape of an ablation lesion caused by ablation in arrhythmia treatment in a subject; and a method for outputting data indicating the shape of an ablation lesion caused by ablation in arrhythmia treatment in a subject.

[0006] Recently, the inventors of the present invention have reported that variables containing information from both active and passive indices provide a better correlation with the shape of the ablation lesion. Therefore, we believe that using real-time changes in all parameters (both active and passive indices) within the ablation system will enable more accurate prediction of the ablation effect and safety.

[0007] The ablation system records real-time changes in numerous active and passive indices over a unit time period. We analyzed the time-series data for these variables during electrical current application using machine learning and investigated their correlation with the shape of the lesions actually formed (specifically, size, surface area (calculated from the major and minor diameters of the surface), depth, maximum major diameter, and volume).

[0008] As a result, the algorithm for predicting lesion shape (specifically, size, surface area (calculated from the major and minor axes of the surface), depth, maximum major axis, and volume) according to the present invention was significantly superior to predictions based on conventional parameters. This algorithm, which uses machine learning to accurately predict ablation results from a large amount of time-series data recorded within the ablation system, is a new technology not previously used in the field of catheter ablation.

[0009] In the future, catheter-based treatments for arrhythmias may further advance, potentially utilizing energies other than radiofrequency. Even with different energies (e.g., pulsed field ablation using electroporation), algorithms for predicting the shape of ablation lesions can be constructed using the method of the present invention by using real-time changing parameters stored within the ablation system. Therefore, it is anticipated that the present invention will remain clinically applicable in the future.

[0010] That is, the present invention provides the following. [Aspect 1] A program for outputting data indicating the shape of a lesion caused by ablation in arrhythmia treatment in a subject, the program being generated by a method including the steps of inputting multiple pieces of data indicating an active or passive indicator of ablation in arrhythmia treatment and the shape of a lesion caused by ablation in arrhythmia treatment into an artificial intelligence model as learning data and allowing the artificial intelligence model to learn. [Aspect 2] The program according to Aspect 1, wherein the multiple pieces of data are multiple pieces of data indicating an active indicator of ablation in arrhythmia treatment, a passive indicator of ablation in arrhythmia treatment, and the shape of a lesion caused by ablation in arrhythmia treatment. [Aspect 3] The program according to Aspect 1, wherein the shape of the ablation lesion is selected from the group consisting of size of the ablation lesion, depth of the ablation lesion, maximum diameter of the ablation lesion, surface area of ​​the ablation lesion, and volume of the ablation lesion. [Aspect 4] The program according to Aspect 1, wherein the ablation lesion is a myocardial lesion. [Aspect 5] The program according to Aspect 1, wherein the active indicator of ablation in arrhythmia treatment is selected from the group consisting of ablation output, contact force, ablation time, catheter contact direction, and type of ablation reflux fluid. [Aspect 6] The program according to Aspect 5, wherein the ablation output is selected from the group consisting of a time integral value of output and an average value of output. [Aspect 7] The program according to Aspect 5, wherein the contact force is selected from the group consisting of an average value of catheter pressing force, a maximum value of catheter pressing force, and a minimum value of catheter pressing force. [Aspect 8] The program according to Aspect 5, wherein the catheter contact direction is selected from the group consisting of a contact angle, a perpendicular direction, and a parallel direction. [Aspect 9] The program according to Aspect 5, wherein the type of ablation reflux fluid is selected from the group consisting of saline and semi-saline. [Aspect 10] The program according to Aspect 2, wherein the passive indicator of ablation in arrhythmia treatment is selected from the group consisting of a local temperature of the ablation site and a local resistance value. [Aspect 11] The program according to Aspect 10, wherein the local temperature of the cauterization site is selected from the group consisting of an average temperature and a maximum temperature.[Aspect 12] The program according to Aspect 10, wherein the local resistance value is selected from the group consisting of a maximum resistance value, a minimum resistance value, a width of decrease in resistance value, and a rate of decrease in resistance value. [Aspect 13] The program according to Aspect 1, wherein the algorithm used in the step of training the artificial intelligence model is a decision tree-based algorithm. [Aspect 14] The program according to Aspect 13, wherein the decision tree-based algorithm is a gradient boosting decision tree-based algorithm. [Aspect 15] A computer storing the program according to Aspect 1. [Aspect 16] A computer-readable recording medium storing the program according to Aspect 1. [Aspect 17] An apparatus for outputting data indicating the shape of an ablation lesion caused by ablation in arrhythmia treatment in a subject, comprising: means for inputting data indicating an active index or a passive index of ablation in arrhythmia treatment in the subject; means for calculating data indicating the shape of an ablation lesion caused by ablation in arrhythmia treatment in the subject using the input data indicating the active index or the passive index of ablation in arrhythmia treatment in the subject and the program described in Aspect 1; and means for outputting the calculated data indicating the shape of an ablation lesion caused by ablation in arrhythmia treatment in the subject. [Aspect 18] The apparatus described in Aspect 17, including means for storing the program described in Aspect 1. [Aspect 19] A method for outputting data indicating the shape of an ablation lesion caused by ablation in arrhythmia treatment in a subject, the method comprising the steps of: inputting data indicating an active or passive index of ablation in arrhythmia treatment in the subject into the device described in Aspect 17; causing the device to calculate data indicating the shape of an ablation lesion caused by ablation in arrhythmia treatment in the subject using the input data indicating the active or passive index of ablation in arrhythmia treatment in the subject and the program described in Aspect 1; and causing the device to output the calculated data indicating the shape of an ablation lesion caused by ablation in arrhythmia treatment in the subject.[Aspect 20] The method of Aspect 19, further comprising the step of acquiring data indicative of active or passive indicators of ablation in arrhythmia treatment in the subject before the step of inputting the data indicative of active or passive indicators of ablation in arrhythmia treatment in the subject into the device of Aspect 17. [Aspect 21] A method for predicting the shape of an ablation lesion due to ablation in arrhythmia treatment, the method comprising the method of Aspect 19.

[0011] According to the program, device, and method of the present invention, data indicating the shape of ablation lesions caused by ablation in arrhythmia treatment in a subject can be output, and the shape of ablation lesions caused by ablation in arrhythmia treatment can be predicted.

[0012] FIG. 1 is a graph showing the correlation between lesion depth and parameters calculated using the statistical analysis software JMP. FIG. 2 is a graph showing the importance of variables for creating an algorithm. FIG. 3 is a graph showing the correlation between lesion surface area and parameters calculated using the statistical analysis software JMP. FIG. 4 is a graph showing the importance of variables for creating an algorithm. FIG. 5 is a graph showing the correlation between lesion volume and parameters calculated using the statistical analysis software JMP. FIG. 6 is a graph showing the importance of variables for creating an algorithm. FIG. 7 is a schematic diagram of a computer that can be used to implement aspects of the present invention. FIG. 8 is a flowchart showing an example of a processing procedure performed in the device and method of the present invention.

[0013] Program: The present invention provides a program for outputting data indicating the shape of ablation lesions caused by ablation in arrhythmia treatment in a subject, the program being generated by a method including the steps of inputting a plurality of data indicating active or passive indicators of ablation in arrhythmia treatment and the shape of ablation lesions caused by ablation in arrhythmia treatment into an artificial intelligence model as learning data and having the artificial intelligence model learn the data.

[0014] Examples of ablation in arrhythmia treatment include high-frequency catheter ablation in arrhythmia treatment, as well as ablation using pulsed field ablation, which utilizes the principle of electroporation in arrhythmia treatment. Preferably, high-frequency catheter ablation in arrhythmia treatment is used.

[0015] The data indicating the shape of the ablation lesion caused by ablation in arrhythmia treatment may be numerical data indicating the shape of the ablation lesion caused by ablation in arrhythmia treatment, or may be symbolic data indicating the shape of the ablation lesion caused by ablation in arrhythmia treatment. The numerical value indicating the shape of the ablation lesion caused by ablation in arrhythmia treatment may be a numerical value indicating the actual shape of the ablation lesion, or may be a numerical value converted based on a specific index.

[0016] The data indicative of an active or passive indicator of ablation in arrhythmia treatment may be a feature vector generated based on the data indicative of an active or passive indicator of ablation in arrhythmia treatment.

[0017] The number of multiple data indicating active or passive indicators of ablation in arrhythmia treatment and the shape of ablation lesions caused by ablation in arrhythmia treatment is preferably 10 to 100,000, more preferably 100 to 10,000, and even more preferably 500 to 5,000.

[0018] There are multiple pieces of data indicating the active or passive indicators of ablation in arrhythmia treatment and the shape of the ablation lesion obtained by ablation in arrhythmia treatment, which means, for example, there are multiple pieces of data indicating, as a set, the active or passive indicators when ablation in arrhythmia treatment is performed and the shape of the ablation lesion obtained by the ablation in arrhythmia treatment.

[0019] The plurality of data indicating active or passive indicators of ablation in arrhythmia treatment and the shape of the ablation lesion caused by ablation in arrhythmia treatment are preferably a plurality of data indicating active indicators of ablation in arrhythmia treatment, passive indicators of ablation in arrhythmia treatment, and the shape of the ablation lesion caused by ablation in arrhythmia treatment.

[0020] There are multiple pieces of data indicating the active indicators of ablation in arrhythmia treatment, the passive indicators of ablation in arrhythmia treatment, and the shape of the ablation lesion obtained by ablation in arrhythmia treatment, which means, for example, there are multiple pieces of data indicating, as a set, the active indicators when ablation in arrhythmia treatment is performed, the passive indicators when ablation in arrhythmia treatment is performed, and the shape of the ablation lesion obtained by the ablation in arrhythmia treatment.

[0021] The shape of the ablation lesion is preferably 1, 2, 3, 4, or 5 selected from the group consisting of the size of the ablation lesion, the depth of the ablation lesion, the maximum major axis of the ablation lesion, the surface area of ​​the ablation lesion, and the volume of the ablation lesion.

[0022] The lesion is preferably a myocardial lesion.

[0023] The active indicators of ablation in arrhythmia treatment are preferably one, two, three, four, or five selected from the group consisting of ablation output, contact force, ablation time, catheter contact direction, and type of ablation reflux fluid. Here, the number refers to the number selected from the categories of ablation output, contact force, ablation time, catheter contact direction, and type of ablation reflux fluid. Therefore, for example, when the number selected from the ablation output is two or more, the number is not limited to these.

[0024] The ablation output is preferably one or two selected from the group consisting of a time integral value of the output and an average value of the output. Here, the ablation output being selected from the group consisting of a time integral value of the output and an average value of the output encompasses the ablation output being a time integral value of the output and an average value of the output.

[0025] The contact force is preferably one, two, or three values ​​selected from the group consisting of the average value of the force pressing the catheter against the target, the maximum value of the force pressing the catheter against the target, and the minimum value of the force pressing the catheter against the target. Here, the number refers to the number selected from the categories of the average value of the force pressing the catheter against the target, the maximum value of the force pressing the catheter against the target, and the minimum value of the force pressing the catheter against the target. Therefore, for example, if the number selected from the average value of the force pressing the catheter against the target is two or more, there is no limitation to these numbers.

[0026] The catheter contact direction may be, for example, one, two, or three selected from the group consisting of a contact angle, a perpendicular direction, and a parallel direction. Here, the catheter contact direction being selected from the group consisting of a contact angle, a perpendicular direction, and a parallel direction includes the catheter contact direction being a contact angle, a perpendicular direction, and a parallel direction. The contact angle may be, for example, a real-time contact angle during ablation.

[0027] The type of ablation perfusion fluid is, for example, one or two selected from saline and semi-saline. Here, the type of ablation perfusion fluid selected from saline and semi-saline encompasses the type of ablation perfusion fluid being saline and semi-saline.

[0028] The passive indicators of ablation in arrhythmia treatment are preferably one or two selected from the group consisting of the local temperature of the ablation site and the local resistance value. Here, the number refers to the number selected from the categories of the local temperature of the ablation site and the local resistance value. Therefore, for example, when the number selected from the local temperature of the ablation site is two or more, the number is not limited to these.

[0029] The local temperature of the cauterization site is preferably one or two selected from the group consisting of an average temperature and a maximum temperature. Here, the local temperature of the cauterization site being selected from the group consisting of an average temperature and a maximum temperature includes the local temperature of the cauterization site being the average temperature and the maximum temperature.

[0030] The local resistance value is preferably one, two, three, or four values ​​selected from the group consisting of the maximum resistance value, the minimum resistance value, the range of decrease in resistance value, and the rate of decrease in resistance value. Here, the number refers to the number selected from the categories of the maximum resistance value, the minimum resistance value, the range of decrease in resistance value, and the rate of decrease in resistance value. Therefore, for example, when the number selected from the maximum resistance value is two or more, there is no limitation to these numbers.

[0031] The generation of a program by a method including a step of inputting training data into an artificial intelligence model and training the artificial intelligence model can be performed by, for example, a known method, but the algorithm used in the step of training the artificial intelligence model is preferably a decision tree-based algorithm. The decision tree-based algorithm is preferably a gradient boosting decision tree-based algorithm such as XGBoost or LightGBM, and more preferably XGBoost.

[0032] The present invention provides a computer that stores the above program, and a computer-readable recording medium that stores the above program.

[0033] Apparatus: The present invention provides an apparatus for outputting data indicating the shape of an ablation lesion caused by ablation in arrhythmia treatment in a subject, the apparatus comprising: means for inputting data indicating an active or passive indicator of ablation in arrhythmia treatment in the subject; means for calculating data indicating the shape of an ablation lesion caused by ablation in arrhythmia treatment in the subject using the input data indicating the active or passive indicator of ablation in arrhythmia treatment in the subject and the program; and means for outputting the calculated data indicating the shape of an ablation lesion caused by ablation in arrhythmia treatment in the subject.

[0034] Examples of ablation in arrhythmia treatment include radiofrequency catheter ablation in arrhythmia treatment and ablation by electroporation in arrhythmia treatment, with radiofrequency catheter ablation in arrhythmia treatment being preferred.

[0035] The data indicating the shape of the ablation lesion caused by ablation in arrhythmia treatment in the subject may be numerical data indicating the shape of the ablation lesion caused by ablation in arrhythmia treatment, or may be symbolic data indicating the shape of the ablation lesion caused by ablation in arrhythmia treatment.The numerical value indicating the shape of the ablation lesion caused by ablation in arrhythmia treatment in the subject may be a numerical value indicating the actual shape of the ablation lesion, or may be a numerical value converted based on a specific index.

[0036] The means for inputting data indicative of an active index or a passive index of ablation in arrhythmia treatment in a subject may be, for example, a means for inputting data indicative of an active index or a passive index of ablation in arrhythmia treatment in a subject. Note that the data indicative of an active index or a passive index of ablation in arrhythmia treatment in a subject may be a feature vector generated based on the data indicative of an active index or a passive index of ablation in arrhythmia treatment in a subject.

[0037] Examples of means for inputting data indicating active or passive indicators of ablation in treating arrhythmia in a subject include a keyboard, a mouse, and the like.

[0038] The means for calculating data indicating the shape of the ablation lesion in the subject due to ablation in arrhythmia treatment using the input data indicating the active or passive indicators of ablation in arrhythmia treatment and the program includes, for example, a central processing unit (CPU) connected to a keyboard, mouse, etc. for inputting the data, connected to a hard disk, flash memory, etc. that stores the program, and connected to memory (storage means) such as ROM and RAM.

[0039] Examples of means for outputting data indicating the calculated shape of ablation lesions in a subject due to ablation in arrhythmia treatment include a monitor, a printer, etc. Also, examples of means for outputting data indicating the calculated shape of ablation lesions in a subject due to ablation in arrhythmia treatment include means for storing the calculated shape of ablation lesions in a subject due to ablation in arrhythmia treatment in a hard disk, flash memory, ROM, RAM, etc.

[0040] The device of the present invention may include a means for storing the program. Examples of the means for storing the program include a hard disk and a flash memory. However, the device of the present invention does not necessarily include the means for storing the program. For example, the device of the present invention may not include the means for storing the program, but may be connected via a communication line to a device that includes the means for storing the program. In other words, the device of the present invention may be part of a system obtained by connecting via a communication line to a device that includes the means for storing the program.

[0041] The device of the present invention may include a means for determining whether to continue ablation based on the output data indicating the shape of the ablation lesion in the subject due to ablation in arrhythmia treatment. Such a means may include, for example, a CPU connected to a hard disk or flash memory that stores a program including conditions for determining whether to continue ablation, and also connected to memories such as ROM and RAM.

[0042] The device of the present invention may include a means for inputting data indicating active or passive indices of ablation in the subject's arrhythmia treatment up to that point, based on the determination by the means for determining whether to continue ablation that ablation should be continued. Examples of such a means include a CPU connected to a hard disk, flash memory, or the like that stores a program including conditions for determining whether to continue ablation, and connected to a hard disk, flash memory, or the like that stores data indicating active or passive indices of ablation in the subject's arrhythmia treatment up to that point, and connected to memory such as ROM and RAM.

[0043] Fig. 7 is a schematic diagram showing an exemplary embodiment of the device of the present invention. In Fig. 7, reference numeral 100 denotes a computer, which includes a control unit 101, a storage unit 102, a peripheral device I / F unit 103, an input unit 104, a display unit 105, and a communication unit 106, all of which are connected by a bus 110. Note that this configuration is merely an example, and various other configurations can be adopted as appropriate.

[0044] The control unit 101 is composed of a CPU (Central Processing Unit), ROM (Read Only Memory), RAM (Random Access Memory), etc. The CPU loads programs stored in the memory unit 102, ROM, recording medium, etc. into a work memory area on the RAM and executes them, driving and controlling each device connected via the bus 110 and realizing the processing performed by the computer. The ROM is a non-volatile memory and stores programs such as the boot program and BIOS of the computer 100, data, etc. The RAM is a volatile memory and temporarily stores programs, data, etc. loaded from the memory unit 102, ROM, recording medium, etc., and also has a work area used by the control unit 101 when performing various processing. The memory unit 102 is, for example, a hard disk drive (HDD) and stores the programs executed by the control unit 101 and various other data.

[0045] The peripheral device I / F (interface) unit 103 is a port for connecting the computer 100 to peripheral devices. The peripheral device I / F unit 103 is configured with a USB, IEEE1394, RS-232C, or the like. The connection with the peripheral devices may be wired or wireless. The input unit 104 has input devices such as a keyboard, a pointing device such as a mouse, and a numeric keypad, and issues operation instructions, operational instructions, data input, and the like to the computer 100. The display unit 105 is a logic circuit or device driver for displaying videos, images, and the like on a display device such as a liquid crystal panel. The input unit 104 and the display unit 105 can also be configured integrally as a touch display.

[0046] The communication unit 106 has a communication control device, a communication port, etc., and is a wired or wireless communication interface that mediates communication with the network 120. The bus 110 is a communication path that mediates the exchange of control signals, data signals, etc. between the devices. The network 120 may further be connected to an external server 130 and a net storage 140.

[0047] FIG. 8 is a flowchart showing an example of a processing procedure performed by the device of the present invention. In this example of the processing procedure performed by the device of the present invention, after the processing procedure starts, time-series ablation data up to the current time is input. Next, a feature vector is generated based on the input time-series ablation data up to the current time. The generated feature vector is input into a trained model, and the ablation lesion size is estimated. Based on the estimated ablation lesion size, it is determined whether to continue ablation. If it is determined that ablation should be continued, time-series ablation data up to the current time is input again, and the same processing procedure as before is repeated. If it is determined that ablation should not be continued, the processing procedure ends.

[0048] Method: The present invention provides a method for outputting data indicating the shape of ablation lesions caused by ablation in arrhythmia treatment in a subject, comprising the steps of: inputting data indicating active or passive indices of ablation in arrhythmia treatment in the subject into the device; calculating data indicating the shape of ablation lesions caused by ablation in arrhythmia treatment in the subject using the input data indicating active or passive indices of ablation in arrhythmia treatment in the subject and the program; and outputting the calculated data indicating the shape of ablation lesions caused by ablation in arrhythmia treatment in the subject from the device.

[0049] Examples of ablation in arrhythmia treatment include radiofrequency catheter ablation in arrhythmia treatment and ablation by electroporation in arrhythmia treatment, with radiofrequency catheter ablation in arrhythmia treatment being preferred.

[0050] The data indicating the shape of the ablation lesion caused by ablation in arrhythmia treatment in the subject may be numerical data indicating the shape of the ablation lesion caused by ablation in arrhythmia treatment, or may be symbolic data indicating the shape of the ablation lesion caused by ablation in arrhythmia treatment.The numerical value indicating the shape of the ablation lesion caused by ablation in arrhythmia treatment in the subject may be a numerical value indicating the actual shape of the ablation lesion, or may be a numerical value converted based on a specific index.

[0051] The step of inputting data indicating active or passive indices of ablation in arrhythmia treatment in a subject into the device is performed using, for example, a keyboard, a mouse, etc. Note that the data indicating active or passive indices of ablation in arrhythmia treatment in a subject may be a feature vector generated based on data indicating active or passive indices of ablation in arrhythmia treatment in a subject.

[0052] The step of inputting data indicating an active indicator or a passive indicator of ablation in arrhythmia treatment in a subject into the device may be, for example, a step of inputting data indicating an active indicator of ablation in arrhythmia treatment in a subject and data indicating a passive indicator of ablation in arrhythmia treatment in the subject into the device.

[0053] The process of having the device calculate data indicating the shape of the ablation lesion caused by ablation in arrhythmia treatment in the subject using the input data indicating the active or passive indicators of ablation in arrhythmia treatment in the subject and the program is carried out by, for example, having a CPU connected to a keyboard, mouse, etc. for inputting the data, connected to a hard disk, flash memory, etc. that stores the program, and connected to memory such as ROM, RAM, etc., perform calculations in accordance with the data and the program.

[0054] The process of outputting data indicating the calculated shape of the ablation lesion caused by ablation in arrhythmia treatment in the subject from the device is performed using, for example, a monitor, printer, hard disk, flash memory, ROM, RAM, etc.

[0055] The method of the present invention may include a step of determining whether to continue ablation based on the output data indicating the shape of the ablation lesion in the subject due to ablation in arrhythmia treatment. This step is performed, for example, by using a CPU connected to a hard disk or flash memory that stores a program including conditions for determining whether to continue ablation, and also connected to memories such as ROM and RAM.

[0056] The method of the present invention may include a step of inputting data indicating active or passive indices of ablation in the subject's arrhythmia treatment up to that point based on the determination to continue ablation in the step of determining whether to continue ablation. This step is performed, for example, using a CPU or the like connected to a hard disk, flash memory, or the like storing a program including conditions for determining whether to continue ablation, a hard disk, flash memory, or the like storing data indicating active or passive indices of ablation in the subject's arrhythmia treatment up to that point, and a memory such as a ROM or RAM.

[0057] The method of the present invention may further include a step of acquiring data indicative of active or passive indicators of ablation in arrhythmia treatment in the subject prior to the step of inputting data indicative of active or passive indicators of ablation in arrhythmia treatment in the subject into the device. Here, the step of acquiring data indicative of active or passive indicators of ablation in arrhythmia treatment in the subject may be, for example, a step of acquiring data indicative of active indicators of ablation in arrhythmia treatment and data indicative of passive indicators of ablation in arrhythmia treatment. The step of acquiring these data is performed, for example, by identifying active or passive indicators expected when ablation in arrhythmia treatment is performed in the subject.

[0058] FIG. 8 is a flowchart showing an example of a processing procedure performed in the method of the present invention. In this example of the processing procedure performed in the method of the present invention, after the processing procedure starts, time-series ablation data up to the current time is input. Next, a feature vector is generated based on the input time-series ablation data up to the current time. The generated feature vector is input into a trained model, and the ablation lesion size is estimated. Based on the estimated ablation lesion size, it is determined whether to continue ablation. If it is determined that ablation should be continued, time-series ablation data up to the current time is input again, and the same processing procedure as before is repeated. If it is determined that ablation should not be continued, the processing procedure ends.

[0059] The present invention provides a method for predicting the shape of an ablation lesion caused by ablation in arrhythmia treatment, the method including the above-described method. When predicting the shape of an ablation lesion caused by ablation in arrhythmia treatment, the shape of the ablation lesion caused by ablation in arrhythmia treatment may be predicted based solely on the meaning of data indicating the shape of the ablation lesion caused by ablation in a subject output by the method of the present invention, or the shape of the ablation lesion caused by ablation in arrhythmia treatment may be predicted by further taking into account the meaning of data indicating the shape of the ablation lesion caused by ablation in a subject output by the method of the present invention.

[0060] Ablation Data Collection: We performed ex-vivo ablation experiments on porcine myocardium using a water circulation system, matching the resistance in the water bath to that observed during clinical ablation. Ablation parameters were set (ablation settings) including ablation power (30 W, 35 W, 40 W, 45 W, or 50 W), contact force (10 g or 20 g), catheter contact direction (perpendicular or parallel), ablation time (10, 20, 30, 60, 120, or 180 seconds), and type of ablation fluid (saline or semi-saline). The depth, surface area, and volume of the myocardial lesion were measured and correlated with each parameter. The correlations with each parameter were compared with the prediction algorithm obtained using machine learning. Of the 1,227 ablation results, 85 lesions were excluded due to steam pop, making it impossible to measure the lesion. The 1,142 lesions that could be measured were analyzed. The data within the ablation system that could be used for the prediction were ablation time, ablation energy (time integral of output), average output, average temperature, maximum temperature, maximum cardiac resistance, minimum cardiac resistance, resistance decrease range, resistance decrease rate, average catheter pressing force, maximum catheter pressing force, and minimum catheter pressing force.

[0061] Methods: The correlation between conventional parameters and bivariate variables for each lesion was calculated using the statistical analysis software JMP. An algorithm called XGBoost was used to calculate predicted values ​​using machine learning. XGBoost is a decision tree-based algorithm that can calculate variable importance, has excellent interpretability, and is a relatively white-box algorithm. Gradient boosting algorithms are known for their high predictive performance and are widely used. In this study, we adopted XGBoost, which had the highest predictive performance among various algorithms.

[0062] As mentioned above, 1,142 data sets were used, and no saline classification was performed. The data set was split into train:test = 75%:25%, and the model's hyperparameters were tuned using a 5-fold cross-validation within the training data. Bayesian optimization was used to tune the parameters, optimizing to minimize the root mean square error (RMSE) between prediction and actual results.

[0063] Results (correlation between depth and parameters): The results are shown in Figures 1 and 2. As shown in Figures 1 and 2, it can be seen that the correlation of the prediction algorithm is significantly higher than that of conventionally used parameters. In terms of the importance of variables for creating the algorithm, it can be seen that ablation time and ablation energy are particularly important. Correlation between the prediction algorithm for Max Depth and actual measured values: RMSE = 0.829 Correlation = 0.907 Correlation**2 = 0.823

[0064] Results (correlation between surface area and parameters): The results are shown in Figures 3 and 4. As shown in Figures 3 and 4, the correlation of the prediction algorithm is significantly higher than that of conventionally used parameters. In terms of the importance of variables for creating the algorithm, it is clear that impedance fluctuations and cauterization output are particularly important. Correlation between the Area prediction algorithm and actual measured values: RMSE = 0.713 Correlation = 0.934 Correlation**2 =0.873

[0065] Results (correlation between lesion volume and parameters): The results are shown in Figures 5 and 6. As shown in Figures 5 and 6, it can be seen that the correlation of the prediction algorithm is significantly higher than that of conventionally used parameters. In terms of the importance of variables for creating the algorithm, it can be seen that ablation energy and ablation time are particularly important. Correlation between volume prediction algorithm and actual measured values: RMSE = 108.580 Correlation = 0.936138852509720 Correlation**2 = 0.876

[0066] Conclusion: This study demonstrated that a prediction algorithm based on machine learning of data from the ablation system can predict lesions more accurately than conventional parameters. It suggests that different parameters are required for predicting lesion depth, surface area, and volume.

Claims

1. A program for outputting data indicating the shape of a lesion caused by ablation in arrhythmia treatment in a subject, the program being generated by a method including the steps of inputting a plurality of data indicating active or passive indicators of ablation in arrhythmia treatment and the shape of a lesion caused by ablation in arrhythmia treatment into an artificial intelligence model as learning data, and having the artificial intelligence model learn the data.

2. The program according to claim 1, wherein the plurality of data are a plurality of data indicating active indicators of ablation in arrhythmia treatment, passive indicators of ablation in arrhythmia treatment, and a shape of ablation lesion caused by ablation in arrhythmia treatment.

3. The program according to claim 1, wherein the shape of the lesion is selected from the group consisting of the size of the lesion, the depth of the lesion, the maximum diameter of the lesion, the surface area of ​​the lesion, and the volume of the lesion.

4. The program according to claim 1, wherein the ablation lesion is a myocardial ablation lesion.

5. The program according to claim 1, wherein the active index of ablation in arrhythmia treatment is selected from the group consisting of ablation output, contact force, ablation time, catheter contact direction, and type of ablation reflux fluid.

6. The program according to claim 5, wherein the ablation output is selected from the group consisting of a time integral value of the output and an average value of the output.

7. The program according to claim 5, wherein the contact force is selected from the group consisting of an average value of the force pressing the catheter against the target, a maximum value of the force pressing the catheter against the target, and a minimum value of the force pressing the catheter against the target.

8. The program of claim 5, wherein the catheter contact direction is selected from the group consisting of a contact angle, a perpendicular direction, and a parallel direction.

9. The program according to claim 5, wherein the type of ablation reflux fluid is selected from the group consisting of saline and semi-saline.

10. The program according to claim 2, wherein the passive indicator of ablation in arrhythmia treatment is selected from the group consisting of local temperature of the ablation site and local resistance value.

11. The program of claim 10, wherein the local temperature of the cauterization site is selected from the group consisting of an average temperature value and a maximum temperature value.

12. The program of claim 10, wherein the local resistance value is selected from the group consisting of a maximum resistance value, a minimum resistance value, a range of resistance value drop, and a rate of resistance value drop.

13. The program of claim 1, wherein the algorithm used in the step of training the artificial intelligence model is a decision tree-based algorithm.

14. The program of claim 13, wherein the decision tree-based algorithm is a gradient boosting decision tree-based algorithm.

15. A computer having stored thereon the program according to claim 1.

16. A computer-readable recording medium storing the program according to claim 1.

17. An apparatus for outputting data indicating the shape of a lesion caused by ablation in arrhythmia treatment in a subject, the apparatus comprising: means for inputting data indicating an active or passive indicator of ablation in arrhythmia treatment in the subject; means for calculating data indicating the shape of a lesion caused by ablation in arrhythmia treatment in the subject using the input data indicating the active or passive indicator of ablation in arrhythmia treatment in the subject and the program described in claim 1; and means for outputting the calculated data indicating the shape of ablation lesion caused by ablation in arrhythmia treatment in the subject.

18. Apparatus according to claim 17, comprising means for storing the program according to claim 1.

19. A method for outputting data indicating the shape of a lesion caused by ablation in arrhythmia treatment in a subject, comprising the steps of: inputting data indicating an active or passive indicator of ablation in arrhythmia treatment in the subject into the device described in claim 17; calculating data indicating the shape of a lesion caused by ablation in arrhythmia treatment in the subject using the input data indicating the active or passive indicator of ablation in arrhythmia treatment in the subject and the program described in claim 1; and outputting the calculated data indicating the shape of ablation lesion caused by ablation in arrhythmia treatment in the subject to the device.

20. The method of claim 19, further comprising the step of acquiring data indicative of active or passive indicators of ablation in treating arrhythmia in a subject prior to the step of inputting the data indicative of active or passive indicators of ablation in treating arrhythmia in a subject into the device of claim 17.

21. A method for predicting the shape of a lesion formed by ablation in arrhythmia treatment, comprising the method according to claim 19.

Images