Apparatus, system and method for identifying subintimal pathways of flexible elongate devices - Patents.com

Abstract

The present invention relates to an apparatus and computer-implemented method for identifying a subintimal path of a flexible elongate device 10 through a lumen, the apparatus including a memory storing data processing instructions configured with curvature related parameters to process data related to the shape and / or position of the elongate device to (i) determine a curvature profile along the elongate device and (ii) associate the curvature profile with a type of curvature indicative of a subintimal path within a lumen, and a processor in communication with the memory and configured to execute the instructions on data related to the shape and / or position measured along at least one segment of the elongate device 10 and output information if a subintimal path is identified.

Description

[Technical field]

[0001] The present invention relates to an apparatus, method and system for identifying a subintimal path of a flexible elongate device through a lumen. The present invention further relates to a corresponding computer program. The present invention is useful in the field of endovascular intervention, in particular for the treatment of chronic total occlusions. [Background technology]

[0002] Chronic total occlusions (CTOs) of arteries can be treated by passing a guidewire through the CTO and then dilating the occluded area. Furthermore, a stent can be deployed. The CTO can be passed intraluminally or subintimally. Physicians are often not aware of exactly how the guidewire passed through the CTO. This can have significant clinical implications. In particular, atherectomy, i.e., debulking of material within the occlusion, can only be performed if the passage of the CTO is intraluminally. Therefore, a detector is needed to detect whether the guidewire is following a subintimal path within the blood vessel.

[0003] US10267624D2 discloses a system for reconstructing the trajectory of an optical fiber. The system includes an optical fiber that is inserted into an object. The optical fiber has a length along which at least one bend sensor unit is disposed. The system also includes a measurement device that measures insertion length increments of the optical fiber, and an interrogation device that detects optical feedback signals from the at least one bend sensor unit. The system further includes a processor device that reconstructs the trajectory of the optical fiber along the inserted length using data pairs based on the measured insertion length increments and the detected optical feedback signals assigned thereto. Summary of the Invention [Problem to be solved by the invention]

[0004] SUMMARY OF THE PRESENT EMBODIMENTS The present invention aims to provide an apparatus that can reliably identify the subintimal path of a flexible, elongated device through a lumen.

[0005] It is a further object of the present invention to provide such an apparatus which warns the user if the elongate device is navigating a subintimal pathway through the blood vessel.

[0006] It is a further object of the present invention to provide a corresponding system and method. [Means for solving the problem]

[0007] In a first aspect of the present invention, there is provided an apparatus for identifying a subintimal path of a flexible elongate device through a lumen, the apparatus comprising: a memory storing data processing instructions configured with parameters related to curvature for processing data related to the shape and / or position of the elongate device so as to (i) determine a curvature profile along the elongate device, and (ii) associate the curvature profile with a type of curvature representative of a subintimal path within the lumen; and a processor in communication with the memory that executes instructions on data related to the measured shape and / or position along at least one segment of the elongate device (10) and outputs information if a subintimal pathway is identified.

[0008] The present invention is based on the idea that a flexible elongate device navigated along a subintimal path should exhibit a shape significantly different from the shape of the flexible elongate device when navigated along an endoluminal path. In particular, the shape of a flexible elongate device navigated along a subintimal path is considered to be a spiral or helical shape. A perfect or ideal helix has a constant curvature along the helix. Thus, determining the curvature profile of the elongate device at one or more shape segments of the elongate device is advantageous for detecting a subintimal path. However, in the subintimal region, the shape of the device may not be a perfect helix, e.g. due to inaccuracies in the shape sensing process, inaccuracies in the data related to shape and / or position, and differences in the actual navigation from a helix (so that the actual shape of the elongate device in the subintimal region deviates from the ideal helix shape). Thus, the present invention proposes to analyze the curvature profile to identify one or more sections in the curvature profile that may be candidates for a subintimal path of the elongate device. This is based on the idea that in the subintimal region of the path of the elongated device, a recognizably higher curvature value should be observed in the curvature profile compared to the intraluminal region of the path of the elongated device. In the case of an intraluminal path, the curvature is zero or small. Thus, according to the present invention, the curvature profile is analyzed to determine whether the elongated device shows high curvature in one or more sections to identify the intraluminal path of the device and output information if a subintimal path is identified. In this way, the present invention allows the occurrence of subintimal navigation of a flexible elongated device to be detected quickly and accurately with a low computational effort. The present invention is beneficial for endovascular interventions, particularly for the treatment of CTOs.

[0009] In a first embodiment, the instructions include: Measuring a curvature profile along at least one geometric segment of an elongate device (10); determining at least one section (A-B) in the curvature profile where the curvature profile exceeds a first curvature threshold level (20) defined as indicative of a deformation of the elongated device (10) that may be due to an obstacle in the path of the elongated device (10); determining a curvature intensity parameter indicative of a curvature intensity in at least one section (A-B) from a curvature value of the curvature profile in at least one section (A-B); identifying a path of the elongate device (10) through the lumen as a subintimal path if the curvature intensity parameter is above a subintimal threshold level; If a subintimal pathway is identified, outputting the above information as a warning.

[0010] In this embodiment, the curvature profile is measured, for example, by a shape sensing modality, and analyzed by a processor to determine at least one section whose curvature profile exceeds a first curvature threshold level defined as representing a deformation of the elongated device that may be due to an obstacle in the path of the elongated device. Other sections whose curvature profile is below the curvature threshold level are ignored to save calculation time. If a section whose curvature profile exceeds the curvature threshold level is determined, a curvature intensity parameter is determined from the curvature value of this section. The curvature intensity parameter may be an indicator of the amount of deviation of the shape from a straight shape. In particular, the curvature intensity parameter may be an indicator of whether the curvature profile in at least one section resembles a curvature profile of a helical or helical shape. If the curvature intensity parameter exceeds a subintimal threshold level defined as representing a subintimal path, the path of the elongated device through the lumen is identified as a subintimal path. If a subintimal path of the elongated device is identified, the physician is alerted.

[0011] During navigation of a flexible, elongated device through a lumen, such as an artery, multiple shape runs are performed and the curvature profiles measured in the shape runs are analyzed as described above to monitor whether subintimal migration is occurring during navigation.

[0012] In a second embodiment, the instructions include: and a curvature classifier configured with parameters related to curvature to receive data related to the shape and / or position of the elongated device and output a classification of the curvature along the at least one segment, the curvature classifier outputting at least one classification of curvature associated with the subintimal path, and a processor in communication with the memory applies the curvature classifier to the data related to the shape and / or position measured along the at least one segment of the elongated device (10) and outputs the above information of the subintimal path if a subintimal path class is found by the curvature classifier.

[0013] In this embodiment, the invention utilizes artificial intelligence by collecting large amounts of data, annotating the data, and using the data to train a deep learning method or another machine learning method (such as a decision tree, random forest, support vector machine, or neural network). An AI-based algorithm can be used to classify the shape into one of the predefined classes, such as tip of the elongated device in a normal region, tip entering the subintimal region, tip passing through the subintimal region, etc., without estimating specific curvature parameters. The input to such an AI-based algorithm can be the shape of the elongated device as a curvature profile of the elongated device or a set of 3D point coordinates, or a combination of the two, allowing the AI-based algorithm to select additional features for the curvature.

[0014] Further embodiments of the present invention are defined below and further disclosed herein.

[0015] In an embodiment, the instructions include determining a curvature intensity parameter as a sum of curvature values ​​in at least one section of the curvature profile. The sum of the curvature values ​​of the curvature profile in the region of interest is then compared to a subintimal threshold level, and if the sum of the curvature values ​​exceeds the subintimal threshold level, a warning is output indicating that a subintimal path has been identified. The term "sum" may also include an integral of the curvature profile in at least one section where the curvature profile exceeds the first curvature threshold level. Summing the discrete curvature values ​​is advantageous in terms of computational cost.

[0016] In another embodiment, the instructions include determining the curvature intensity parameter as an average of the curvature values ​​in at least one section of the curvature profile. Calculating the average of the curvature values ​​is also advantageous in terms of computational cost.

[0017] In a further embodiment, the instructions include identifying at least one section as not indicative of subintimal migration if the maximum curvature value falls below a second curvature threshold level that is higher than the first curvature threshold level. If the maximum curvature value falls below the second curvature threshold level, the at least one section is considered invalid and the determination of the curvature intensity parameter for this section is omitted. Applying the analysis of the curvature profile in the section where the curvature profile is above the first curvature threshold level to a higher second curvature threshold level has the advantage of preventing noisy calculations in this section and discarding curvature profiles that exceed the first threshold level for a long period of time but do not reach a maximum value as high as the second threshold level. The ratio between the second curvature threshold and the first curvature threshold may be in the range of 1.5 to 3 (e.g., 2).

[0018] In a further embodiment, the instructions include determining the at least one section by determining a point at which the curvature profile passes a first curvature threshold level, This operation in analyzing the curvature profile has the advantage of being computationally less expensive.

[0019] The measurement of the curvature profile may be limited to one shape segment of the elongate device. In this case, this shape segment preferably includes the distal tip of the device, since the occurrence of a subintimal passage of a flexible elongate device is most likely at the distal tip of the device. The measurement of the curvature profile may also be performed along one or more other shape segments between the distal tip and the proximal end. A curvature intensity parameter is calculated for each of these shape segments.

[0020] Alternatively, the region for calculating the curvature intensity parameter can be determined from anatomical information. For example, the occluded region can be identified on a contrast-enhanced X-ray or 3D anatomical data. After shape registration to the X-ray and 3D anatomical data, the method according to the invention can be started when the flexible elongated device is in the occluded region identified in the anatomical image data.

[0021] In a further embodiment, the instructions include determining a distance of a distal end of the at least one section from a distal tip of the elongate device and identifying the at least one section as a distal tip portion if the determined distance is less than a threshold distance.

[0022] If the determined distance is greater than the threshold distance, the segment is not considered to be a segment that includes the distal tip of the device, and the curvature strength at the distal tip segment may be set to zero.

[0023] In a further embodiment, the instructions include setting at least two different subintimal thresholds and outputting different warnings depending on the at least two different subintimal thresholds. It is advantageous to provide a warning scheme that takes into account the severity of deviation of the elongated device's movement from the endoluminal path. For example, multiple warning levels are defined and different warnings are output depending on which of these levels the curvature intensity parameter exceeds. For example, color-coded warnings are issued, such as yellow, orange, bright red, and dark red, where the color sequence indicates increasing severity of deviation of the movement from the endoluminal path.

[0024] In a further preferred embodiment, the instructions include setting a time limit and outputting a warning if a curvature profile representative of a subintimal path is detected for a period exceeding the time limit. In this embodiment, outputting the warning depends, for example, not only on exceeding the subintimal threshold but also on a timing constraint. That is, a warning is only output if both the subintimal threshold and the time limit are exceeded. This embodiment takes into account that subintimal movement may only occur for a short period of time (less than 1 or 2 seconds), after which the elongate device may again proceed along an intraluminal path. Thus, a warning can be avoided when it is not necessary.

[0025] In connection with the above embodiment, depending on which of the at least two different subintimal thresholds is defined, different time limits are defined for the different subintimal thresholds, the time limits being adapted to the warning level, e.g., for a low warning level the time limit is set higher than for a high warning level.

[0026] In further embodiments, the alert is output as a tactile, audible, visual or textual alert, the alert may be visualized on a display or played through a speaker.

[0027] If the apparatus is configured to display the shape of the elongated device on a monitor or the like, the color of the visualized shape of a region of interest that has resulted in a high curvature intensity parameter can also be changed, for example from green (movement within the lumen) to red (movement under the intima).

[0028] In a further embodiment, the apparatus includes an optical shape sensing modality that includes an optical interrogation modality that optically interrogates an optical fiber included in the elongate device and receives optical feedback from the optical fiber.

[0029] In addition to optical shape sensing, other 3D shape sensing sources can also be used in the present invention, for example, electromagnetic tracking systems can be used.

[0030] According to a second aspect of the invention there is provided a system comprising an elongated flexible device and an apparatus according to the first aspect.

[0031] The flexible elongate device may be a guidewire or a catheter.

[0032] According to a third aspect of the present invention, there is provided a computer-implemented method for automatically identifying subintimal movement of a flexible elongate device through a lumen, the method comprising: providing data processing instructions configured with curvature related parameters to process data relating to the shape and / or position of the elongate device so as to (i) determine a curvature profile along the elongate device, and (ii) associate the curvature profile with a type of curvature representative of a subintimal path within the lumen; and executing data processing instructions on data relating to the measured shape and / or position along at least one segment of the elongate device and outputting information if a subintimal pathway is identified.

[0033] A first embodiment of the method comprises: providing data representative of a curvature profile along at least one geometric segment of an elongate device; determining at least one section in the curvature profile where the curvature profile exceeds a first curvature threshold level defined as indicative of a deformation of the elongate device that may be due to an obstacle in the path of the elongate device; determining a curvature intensity parameter indicative of a curvature intensity in the at least one section from the curvature values ​​of the curvature profile in the at least one section; identifying a path of the elongate device through the lumen as a subintimal path if the curvature intensity parameter is above a subintimal threshold; and outputting a warning if subintimal migration is identified.

[0034] A second embodiment of the method comprises: providing a curvature classifier, preferably a trained curvature classifier, configured with parameters related to curvature to receive data related to the shape and / or position of the elongate device and to output a classification of the curvature along said at least one segment, wherein the curvature classifier outputs at least one classification of a curvature associated with the subintimal pathway; and applying a curvature classifier to data relating to the measured shape and / or position along said at least one segment of the elongate device and outputting said information of the subintimal pathway if a subintimal pathway class is found by the curvature classifier.

[0035] In a fourth aspect of the present invention there is provided a computer program comprising program code means which, when executed on a computer, causes the computer to carry out the steps of the method according to the third aspect.

[0036] It is to be understood that the claimed systems, methods and computer programs have similar and / or identical preferred embodiments as the claimed apparatuses, in particular as defined in the dependent claims and as disclosed in this specification.

[0037] The present disclosure encompasses a non-transitory computer-readable recording medium having a computer program stored thereon. [Brief description of the drawings]

[0038] These and other aspects of the invention will be apparent from and elucidated with reference to the exemplary embodiments described hereinafter.

[0039] [Figure 1] FIG. 1 shows a schematic diagram illustrating the intraluminal and subintimal passage of a flexible elongate device through the lumen of a blood vessel. [Diagram 2]FIG. 2 shows a schematic of a flexible elongate device spirally wrapped around a ballpoint filler mimicking a flexible elongate device within a subintimal passage. [Figure 3a-3d] 3a-3d show visualized reconstructions of the elongated device of FIG. 2 in various views. [Figure 4] FIG. 4 shows a visualized reconstruction of an elongated device actually inserted into the lumen of a blood vessel. [Diagram 5] FIG. 5 shows a flow diagram of a method for identifying a subintimal pathway for a flexible elongate device. [Figure 6] FIG. 6 shows an illustration of the measured curvature profile of an elongated device under the influence of a subintimal pathway. [Figure 7] FIG. 7 shows diagrams of the curvature profile at different locations of the device along the subintimal pathway. [Figure 8] FIG. 8 shows an illustration of the curvature profile at yet another location of the device along the subintimal pathway. [Figure 9] FIG. 9 shows a diagram of the curvature profile of FIG. 8 illustrating an analysis of the curvature profile of FIG. [Figure 10] FIG. 10 shows a diagram of the curvature profile of FIG. 8 showing further analysis of the curvature profile of FIG. [Figure 11] FIG. 11 shows another view of the curvature profile of FIG. 8 showing further analysis of the curvature profile of FIG. [Figure 12] FIG. 12 shows a diagram illustrating the curvature intensity parameters of the distal shape segment of the elongate device for several shape measurements when no subintimal path occurs during navigation of the elongate device. [Figure 13] FIG. 13 is a view similar to FIG. 12, but with a subintimal pathway occurring during navigation of the device. [Figure 14] FIG. 14 is a block diagram of a system for identifying a subintimal pathway of a flexible elongate device. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0040] FIG. 1 shows a schematic diagram for illustrating the intraluminal and subintimal paths of a flexible elongated device 10 through the lumen of a blood vessel V. The blood vessel V is, for example, an artery. The blood vessel may be affected by a chronic total occlusion (CTO) that narrows the lumen of the blood vessel V in the vessel wall VW. The CTO may be treated by passing the elongated device 10, which may be a guidewire GW, through the CTO. After passing the guidewire GW through the CTO, the occluded area may be dilated. Additionally, a stent (not shown) may be placed. To treat the CTO in an endovascular intervention, the elongated device 10 is navigated through the blood vessel V, i.e., advanced through the lumen of the blood vessel V according to the arrow 14. Reference number 12 indicates a distal portion of the flexible elongated device 10. When navigating the elongated device 10 through the blood vessel V, the elongated device 10 may follow an intraluminal or subintimal path. The intraluminal path of the elongated device 10 is shown in solid lines, and the subintimal path is shown in dashed lines. For proper treatment of the CTO, the elongated device 10 must pass the CTO in an intraluminal path. For example, atherectomy (debulking of material within the occlusion) can only be performed if the passage of the CTO is intraluminal, but passage of the CTO in an intraluminal path is insufficient for proper treatment of the CTO. However, physicians often do not know exactly how the elongated device passed the CTO. The present disclosure provides a computer-implemented method and apparatus for automatically identifying the subintimal path of an elongated flexible device such as the device 10 and providing real-time alerts of the subintimal CTO track, allowing physicians to adapt treatment during surgery and ensure that atherectomy is only utilized if the passage of the CTO by the elongated device 10 is intraluminal. Additionally, the disclosure herein can provide real-time alerts of the extent of the subintimal CTO track.

[0041] With reference to Figures 2-4, we will now describe how the 3D shape of an elongated device such as guidewire GW appears as it navigates the subintimal pathway.

[0042] Bench testing was performed with an optical shape sensing enabled guidewire GW as shown in FIG. 2. In FIG. 2, the guidewire GW is spirally wound around a 3 mm diameter ballpoint filler 16 that mimics the 3 mm lumen of a blood vessel. The present teachings are based on the idea that the guidewire, when in the subintimal pathway, will assume a spiral shape along the subintimal pathway. Thus, the spiral shape of the guidewire GW around the ballpoint filler is made to mimic the expected spiral shape of the guidewire in the subintimal space.

[0043] The guidewire GW spirally wound around the ballpoint pen filler 16 is subjected to shape sensing, here optical shape sensing. Optical shape sensing is performed using an optical frequency domain reflectometry (OFDR) sensing system. Figures 3a) to 3d) show reconstructions of the guidewire in various views created by the optical shape sensing viewing software. Figure 3a) shows a front view of the ballpoint pen filler 16, Figure 3b) shows a side view, Figure 3c) shows an enlarged front view, and Figure 3d) shows a view in the longitudinal direction.

[0044] As is evident from Figures 3a)-3d), the 3D reconstruction of the guidewire has a spiral structure in region A, which is spirally wound around the ballpoint pen filler 16. This differs from the usual straight structure in region B.

[0045] FIG. 4 shows a visualization of the reconstructed shape of an optical shape sensing enabled guidewire, where the guidewire is actually moving subintimally through a CTO. As can be seen from FIG. 4, even in the real case, the guidewire GW has a spiral structure in the subintimal zone region S and a normal / straight structure in the intraluminal region I. This distinction is the basis of the present disclosure for automatically detecting whether the guidewire GW is in the intraluminal space, in the subintimal space, or in a partially subintimal and partially intraluminal path. According to the present disclosure, the transition from the intraluminal space to the subintimal space and vice versa can also be detected.

[0046] Based on the idea that the shape of a flexible elongate device such as a guidewire has a spiral or helical structure in the subintimal region, we further explain how the spiral structure of the elongate device 10 can be detected with minimal computational effort.

[0047] As explained above, the 3D shape of the elongate device 10 in the subintimal passage may be in the form of an (incomplete) helix. A perfect cylindrical helix can be mathematically described by the following parameterization: x(t)=a cos(t) y(t)=a sin(t), z(t)=bt, where a is the radius and b / a is the slope of the helix. Due to the similarity of the 3D shape of the elongated device 10 in the subintimal path with the helix, the distal part of every shape can be fitted to the above parameterization and evaluated for how well it fits the helix. However, this calculation operation requires a large amount of calculations if it is performed for every shape measurement during navigation of the elongated device 10, and the calculations do not always yield a unique solution. Furthermore, the shape of the elongated device does not always fit the helix due to the inaccuracy of the shape measurements and the formation of the actual subintimal path. Therefore, the present teachings propose an indirect but more practical solution that can handle both the incompleteness of the shape sensing data and the subintimal movement without requiring a large amount of calculations.

[0048] The method according to the present disclosure analyzes the curvature profile obtained from measuring the shape of the elongate device 10. The reason that the curvature is suitable for identifying subintimal pathways is because the curvature of a perfect cylindrical helix is ​​a constant, subject to the parameters a and b above: |a| / (a 2 +b 2 )

[0049] Since the shape of the elongate device 10 in the subintimal pathway is expected to resemble a helix shape, the 3D shape of the elongate device 10 is expected to have a significantly higher curvature in the subintimal pathway compared to the endoluminal pathway. If the subintimal pathway yields an ideal helix shape for the elongate device, a rectangular curvature profile with a high constant curvature should be observed. When the distal portion 14 of the device 10 is in the subintimal pathway, the constant curvature profile is high at the most distal portion of the curvature profile.

[0050] In reality, due to shape sensing imprecision, data imprecision, and variance of actual navigation from a perfect helix, the actual shape of the elongated device within the subintimal pathway will deviate from an ideal helix, and therefore the curvature profile will deviate from a perfect rectangular profile, although the curvature profile will remain highly curvatured throughout the entire subintimal pathway.

[0051] 5 shows a flow diagram of an embodiment of a computer-implemented method for automatically identifying a subintimal pathway of an elongate device. The method 100 includes a step S102 of measuring a curvature profile along at least one shape segment of the elongate device 10. The measurement of the curvature profile is performed by evaluating shape sensing data obtained from a shape sensing modality, such as an optical shape sensing modality or any other source of shape sensing data. The curvature is extracted or calculated from the shape sensing data, as known to those skilled in the art.

[0052] FIG. 6 shows a diagram of a curvature profile at a distal portion of the elongated device 10, where the distal portion is in a subintimal pathway. The x-axis of FIG. 6 shows sample points (nodes) of the shape sensing data, and the y-axis shows the curvature (in units of reciprocal sample points). Sample point 350 is the tip of the elongated device 10. The shape sensing data of FIG. 6 was actually obtained when the elongated device was navigated through a blood vessel. As is evident from FIG. 6, the curvature profile has curvature values ​​from sample points 270-350 that are significantly higher compared to the curvature values ​​of the curvature profile below sample point 270. As shown in FIG. 7 and FIG. 8, similar curvature profiles occur at various locations of the device along the subintimal pathway. Thus, as long as the device 10 is in the subintimal pathway, the curvature profile of each shape measurement shows high curvature in the subintimal pathway during navigation of the device 10. As can be seen from Figures 6-8, the curvature values ​​in the regions of higher curvature are different from Figures 6 to 7 and from Figures 7 to 8. Nevertheless, the curvature profiles show distinct regions of higher curvature, i.e., the distal portion of device 10, including the tip (sample point 350), is assumed to be in the subintimal pathway.

[0053] Referring again to FIG. 5, the method 100 proceeds to step S104 of determining at least one section in the curvature profile where the curvature profile exceeds a first curvature threshold level. The curvature threshold level is defined as representing a deformation of the elongated device that may be due to an obstacle in the path of the elongated device. FIG. 9 shows an embodiment of an implementation of step S104 of the method. The first curvature threshold level 20 is preferably set as a low curvature value (such as 0.5) as shown in FIG. 9. The determination of at least one section where the curvature profile exceeds the first curvature threshold level is done by determining points A and B where the curvature profile passes the first curvature threshold level 20. This can then be done by examining the curvature profile from distal (sample point 350) to proximal (0) (or vice versa) to find the positions A and B where the curvature profile passes the level 20.

[0054] 5, the method 100 proceeds to step S106 of determining a curvature intensity parameter indicative of how strong the curvature is in at least one section A-B from the curvature values ​​of the curvature profile in at least one section where the curvature profile exceeds a first curvature threshold level 20. In the embodiment shown in FIG. 10, the curvature intensity parameter is determined as a sum of the curvature values ​​of sections A-B of the curvature profile. The act of summing the curvature values ​​in sections A-B is indicated in FIG. 10 by vertical line 22. In another embodiment, the curvature intensity parameter is determined as an average of the curvature values ​​of sections A-B of the curvature profile.

[0055] 5, the method 100 includes a step S108 of identifying the path of the elongate device 10 through the lumen as a subintimal path if the pre-determined curvature intensity parameter is above a subintimal threshold. If the curvature intensity parameter is determined as the sum of the curvature values ​​in sections A-B, the subintimal threshold may be a value that is set much larger than the first curvature threshold level 20, for example an order of magnitude larger.

[0056] The method 100 further proceeds to step S110 of outputting an alert if subintimal migration is identified. The alert may be output as a tactile, audible, visual, or textual alert.

[0057] Method 100 may further include the step of identifying sections A-B as not indicative of subintimal movement if, for example, before step S106 and after step S104, the maximum curvature value in sections A-B falls below a second curvature threshold level higher than the first curvature threshold level 20. This is shown in FIG. 11. In FIG. 11, a second curvature threshold level 24 is shown. For example, the second curvature threshold level 24 may be in the range of 1.5 to 2.5 times the first curvature threshold level 20, for example 2 times. In the example of FIG. 11, the second curvature threshold level 24 is set to 0.1. It is therefore twice the first curvature threshold level. This step may also be advantageously used to avoid noisy calculations or to discard curvature profiles that exceed the first threshold for a long period of time but do not reach a maximum value as high as the second threshold.

[0058] Although the above description only considers the distal shape segment including the tip of the elongate device 10, other shape segments of the elongate device 10 between the distal and proximal ends may also be considered, but it is most important and advantageous to perform the method on the distal shape segment of the elongate device 10.

[0059] Method 100 may further include, after step S104, determining a distance of distal end A of section A-B from distal tip 14 of elongate device 10, and identifying section A-B as a distal tip portion if the determined distance is less than a threshold distance. Otherwise, the segment under consideration is not considered a distal shape segment, and if the method for identifying a subintimal pathway is applied or directed only to the distal tip portion of device 10, the curvature strength parameter of the distal shape segment is set to zero.

[0060] In summary, a curvature intensity parameter is determined for each shape of the device being navigated, or multiple curvature intensity parameters are determined if, for example, non-distal shape segments are also evaluated, and a decision can be made based on these curvature intensity parameters. Figures 12 and 13 show the distribution of these curvature intensity parameters for a number of shape runs. Figure 12 shows the determined curvature intensity parameters for more than 10,000 shape measurements (shape runs) recorded without a subintimal path, and Figure 13 shows one with a subintimal path. The curvature intensity parameters are calculated without requiring the maximum curvature value in sections A-B to exceed the second curvature threshold level 24 (Figure 11). Requiring the maximum curvature value in sections A-B to exceed the second curvature threshold would affect some low values ​​in Figures 12 and 13, but not the high values ​​that are more important for whether to issue a warning.

[0061] In FIG. 12, all curvature intensity parameters determined for each shape run are below the subintimal threshold of 4, with most values ​​below 2, indicating successful navigation of the elongated device.

[0062] 13, the curvature intensity parameter for each shape run shows subintimal movement after shape run 10,000 and a short trial around shape run 3,500. Starting with shape run 10,000 the values ​​are mostly above 6 indicating a subintimal path for the elongated device 10, but in the region of navigation indicating an endoluminal path the curvature intensity parameter falls below 4.

[0063] The method of the present teachings also contemplates setting more than one subintimal threshold, such as two, three, four, five, or more, and outputting different warning levels in response to different subintimal thresholds. For example, multiple warning levels may be pre-set, such as four, six, eight, ten, etc., and color-coded warnings, such as yellow, orange, light red, and dark red, are generated when the curvature intensity parameter exceeds any of these levels. Thus, the severity of deviation of the path of the elongate device 10 from the intraluminal path can be indicated to the physician.

[0064] It is also possible within the methods of the present teachings to include timing constraints on the alert levels; that is, one or more time limits can be set. If there are multiple different time limits, each time limit can be assigned to one of the different subintimal thresholds. For example, a 2 second time limit can be assigned to a curvature intensity parameter below 6 and a 1 second time limit can be assigned to values ​​above 6. An alert is provided only if both the amplitude and timing constraints are met.

[0065] Additionally, the color of the visualized shape of the region of interest that resulted in a high curvature intensity parameter may be changed, for example from green to red.

[0066] Another embodiment of a method for identifying the subintimal pathway is described below.

[0067] The above curvature-related parameters can be estimated using supervised machine learning and optimization techniques. The same methods as above can be used, except that the above curvature-related parameters are optimally determined using an annotated set of shapes. The annotated shape set includes the following for each shape in the set: -Geometry information (position, distortion, curvature, twist, alpha value, etc.), - high curvature start and end segments, - the first label of the whole shape indicating whether the tip of the shape is in the CTO region, - A second label for the entire shape indicating the degree to which the shape has followed a subintimal path, where the degree may be a continuous measurement or a level from a multi-level scale (e.g., binary, with 0 being a special case where 0 indicates an intraluminal path and 1 indicates a subintimal path).

[0068] The annotated set can be used to determine values ​​for curvature-related parameters, such as first and second curvature thresholds, to maximize detection of shape segments with high curvature. Since there are only two parameters to estimate, an exhaustive search of combinations is possible, and the accuracy can be evaluated using the first labels of the annotated dataset above, and the combination of values ​​that results in the highest accuracy can be selected.

[0069] A similar method can be used to determine the value of the subintimal threshold of the curvature intensity. In this case, the second label can be used and the accuracy can be determined by maximizing it. This optimal threshold can be calculated separately using the sum or average of the curvatures in the determined shape segments. If another method of measuring the curvature intensity is conceived, that method can be used to recalculate the optimal threshold.

[0070] Described below is another embodiment of a method for identifying subintimal pathways using AI-based shape classification.

[0071] In this embodiment, the likelihood of subintimal migration is used to indicate the extent to which the device passes the CTO subintimally. In the previous embodiment, this was determined as 0 / 1 by comparing the curvature intensity to a threshold. In the AI-based method, the curvature intensity can be calculated as a continuous value between 0 and 1 without explicit use of the curvature intensity. Thus, the likelihood of subintimal migration is a more general term to describe this situation.

[0072] AI-based shape classification can be used to:

[0073] The entire shape or a portion of it can be classified as subintimal or intraluminal, and in addition, a degree can be assigned, which indicates the likelihood of subintimal migration. On the scale of [0,1], a low value indicates that the shape or a portion of it is more likely to follow an intraluminal route, and a high value indicates a subintimal route.

[0074] Training a deep neural network to classify shapes as subintimal or endoluminal pathways can be achieved by using the same annotated set as above. In this case, the deep neural network can be trained by using the shape as input and the label (subintimal or endoluminal) as the ground truth class. Shape features such as position, curvature, distortion, twist, alpha value can be provided as input. It is also possible to provide multiple shapes as input. The network architecture can be a classification network. Typical architectures found in the literature can be used for this purpose, such as the ResNet architecture (K He, X Zhang, S Ren, J Sun, "Deep residual learning for image recognition") (Computer Vision and Pattern Recognition Conf., 2015).

[0075] In another embodiment, a deep neural network can be trained to classify shapes into different classes along their length. For this task, the same dataset as above can be used, but the ground truth is generated using information about the start and end segments of shapes with high curvature. For every shape, a ground truth shape consisting of 0s and 1s is generated. Values ​​of 0 are assigned to the positions of the shape outside the CTO and values ​​of 1 are assigned to the positions inside the CTO. Given a shape of length N samples (points) as input, the ground truth for that shape will also be N samples long, with values ​​of 0 outside the CTO and values ​​of 1 inside the CTO. This type of problem is also known as a segmentation problem. As mentioned above, state-of-the-art neural networks can be used for shape segmentation. A common segmentation network is U-Net (O. Ronneberger, P. Fischer, T. Brox, U-Net: Convolutional Networks for Biomedical Image Segmentation, Medical Image Computing and Computer-Assisted Intervention (MICCAI) (Springer, LNCS, Vol. 9351:234-241, 2015). Again, as in classification, shapes with features such as position, curvature, twist, distortion, etc. can be used as input, and a segmented shape can be estimated as output. This output segmentation map can be compared to the ground truth shape segmentation during training of the deep neural network. Once sufficient accuracy is achieved, or progress between iterations slows down, training is stopped and the best network during these iterations is used to segment the shape. Using this method, parts of the shape that pass through the CTO or healthy segments can be identified.

[0076] FIG. 14 illustrates a system 50 for identifying a subintimal pathway for an elongate device 10 for performing a method according to the present teachings.

[0077] The system 50 comprises a device 52 including a shape sensing modality 54, a memory 56, and a processor 58. The device may further comprise a display and / or a speaker (not shown).

[0078] The shape sensing modality 54 measures the curvature profile according to step S102. The memory 56 stores data processing instructions configured with curvature related parameters for processing data related to the shape and / or position of an elongated device such as device 10. The shape and / or position related data can be used to determine a curvature profile along the elongated device and associate the curvature profile with a type of curvature indicative of a subintimal pathway within the lumen. A processor 57 is in communication with the memory 56 and executes instructions on the shape and / or position related data measured along at least one segment of the elongated device (10) and outputs information if a subintimal pathway is identified.

[0079] Processor 58 executes steps S104 to S110.

[0080] The shape sensing modality 54 may be an optical shape sensing modality, including an optical interrogation modality that optically interrogates an optical fiber 60 included in the elongated device 10 and receives optical feedback from the optical fiber 60. The system 50 further includes the elongated, flexible device 10, which may be configured, for example, as a guidewire.

[0081] Data processor 58 may be implemented in hardware, firmware, or software.

[0082] While the invention has been illustrated and described in detail in the drawings and the foregoing description, such illustration and description are to be considered illustrative or exemplary and not restrictive. The invention is not limited to the disclosed embodiments. Other variations of the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure, and the appended claims.

[0083] In the claims, the word "comprising" does not exclude other elements or steps, and the singular elements do not exclude a plurality. A single element or other modality may fulfill the functions of several items recited in the claims. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage.

[0084] The computer program comprising the code means for carrying out the method according to the invention can be stored / distributed on a suitable non-transitory medium, such as an optical storage medium or a solid-state medium, supplied together with or as part of other hardware, but can also be distributed in other forms, such as via the Internet or other wired or wireless communication systems.

[0085] Any reference signs in the claims should not be construed as limiting the scope.

Claims

1. An apparatus for identifying a subintimal pathway of a flexible elongated device passing through a lumen, comprising: a memory storing data processing instructions configured with parameters related to curvature for processing data related to the shape and / or position of the elongated device so as to (i) determine a curvature profile along the elongated device, (ii) associate the curvature profile with a type of curvature representing a subintimal pathway within the lumen, and (iii) output information when the subintimal pathway is identified; a processor communicating with the memory and executing the data processing instructions on data related to the shape and / or position measured along at least one segment of the elongated device and outputting information when the subintimal pathway is identified; The apparatus comprising the above.

2. The step (i) of the data processing instructions includes measuring a curvature profile along at least one shape segment of the elongated device The step (ii) of the data processing instructions includes determining, in the curvature profile, at least one section exceeding a first curvature threshold level defined as representing a deformation of the elongated device that may be due to an obstacle in the path of the elongated device; determining a curvature intensity parameter indicating the curvature intensity in the at least one section from the curvature values of the curvature profile in the at least one section; identifying the path of the elongated device passing through the lumen as a subintimal pathway when the curvature intensity parameter exceeds a subintimal threshold level; outputting the information as a warning when the subintimal pathway is identified; The apparatus according to claim 1, comprising the above.

3. The data processing instructions are ​ Performing steps (i) and (ii) of the data processing instructions, receiving data related to the shape and / or position of the elongated device, and including a curvature classifier composed of parameters related to curvature to output a classification of the curvature along the at least one segment, the curvature classifier outputting at least one classification of the curvature associated with the subintimal path, the processor communicating with the memory applying the curvature classifier to data related to the shape and / or position measured along the at least one segment of the elongated device, and outputting the information of the subintimal path when the subintimal path class is found by the curvature classifier. The device according to claim 1.

4. Determining the curvature strength parameter includes calculating the sum of the curvature values in the at least one section of the curvature profile or calculating the average of the curvature values in the at least one section of the curvature profile. The device according to claim 2.

5. The step (ii) of the data processing instructions further includes, after a sub-step of determining a curvature strength parameter from the curvature values of the curvature profile in the at least one section, identifying that the at least one section does not indicate subintimal movement when the maximum curvature value among the determined curvature values is below a second curvature threshold level that is higher than the first curvature threshold level. The device according to claim 2.

6. The step (ii) of the data processing instructions further includes determining the at least one section by determining a point at which the curvature profile passes through the first curvature threshold level before a sub-step of determining a curvature strength parameter from the curvature values of the curvature profile in the at least one section. The device according to claim 2.

7. The at least one shape segment of the elongated device includes the distal tip of the elongated device. The device according to claim 1.

8. Step (ii) of the data processing instruction further includes determining the distance from the distal end of the at least one section of the elongated device to the distal end of the at least one section from the curvature value of the curvature profile in the at least one section, and if the determined distance is less than a threshold distance, identifying the at least one section as the distal tip portion, before determining the curvature strength parameter from the curvature value of the curvature profile in the at least one section. The apparatus according to claim 7.

9. The data processing instruction includes setting at least two different subendocardial thresholds and outputting different warnings according to the at least two different subendocardial thresholds. The apparatus according to claim 2.

10. Step (ii) of the data processing instruction further includes setting a time limit, and after the substep of determining the curvature strength parameter from the curvature value of the curvature profile in the at least one section, if the curvature profile representing the subendocardial path is detected for a period exceeding the time limit, outputting the information. The apparatus according to claim 1.

11. The data processing instruction includes setting different time limits, and each time limit is assigned to any one of the at least two different subendocardial thresholds. The apparatus according to claim 9 and claim 10.

12. The data processing instruction includes outputting the information as tactile information, audible information, visual information, or text information. The apparatus according to claim 1.

13. An elongated flexible device, The apparatus according to any one of claims 1 to 12, A system comprising.

14. A computer-implemented method for automatically identifying a subendocardial path of a flexible elongated device passing through a lumen, comprising: (i) determining a curvature profile along the elongated device, and (ii) providing data processing instructions composed of parameters related to curvature to process data related to the shape and / or position of the elongated device so as to associate the curvature profile with a type of curvature representing a subendocardial path within the lumen; Executing the data processing instruction on data related to the shape and / or position measured along at least one segment of the elongated device, and outputting information when the subendocardial path is identified; A computer-implemented method comprising.

15. A computer program including program code means for causing a computer to execute the steps of the computer-implemented method according to claim 14 when executed on a computer.

Images