Determining the end position of the stent

Abstract

The system 100 for determining suitability of a location along a lumen 110 as an end location for a stent includes acquiring at least a first endoluminal image 140 from an endoluminal imaging device 130. i⊂1..n endoluminal image 140 representing a position along lumen 110, 1..n receiving a first endoluminal image 140 from the first endoluminal image receiving section 110; i⊂1..n represents a suitable end location for the stent.

Description

[Technical field]

[0001] The present disclosure relates to determining the suitability of a location along a lumen as an end point location for a stent. A system, computer-implemented method, and associated computer program product are disclosed. [Background technology]

[0002] In the medical field, stents are frequently inserted into blood vessels, coronary arteries, biliary, bronchial, urinary, and other lumens within the anatomy to keep the relevant passageways open, with the need and ultimate location of the stent being determined by the physician, often under the guidance of an intraluminal imaging device.

[0003] As an example, deep vein intervention typically begins with the generation of an initial venogram of the patient's vasculature using contrast-enhanced x-ray imaging. A guidewire is then inserted into the vasculature under x-ray imaging, and an intraluminal imaging device, such as an intravascular ultrasound "IVUS" imaging catheter or an optical coherence tomography "OCT" imaging catheter, is translated along the guidewire in a "pull-back" procedure to image and thereby diagnose portions of the lumen within the vasculature. Stenosis, or "compression," or "thrombosis," within the lumen that restricts blood flow often indicates vascular disease, whereas more open areas of the lumen, where blood flow is less restricted, typically indicate healthy areas. When a stent is deemed necessary, for example when the lumen is compressed by more than 50% relative to a baseline healthy area, the physician may determine the appropriate location, or "landing zone," of the stent from the IVUS image by identifying diseased and healthy areas of the lumen. The physician may also identify bifurcations, also known as "confluences," within the lumen from the IVUS image. In vascular procedures, physicians typically designate a landing zone for the stent so that the stent overlaps the diseased area, the ends of the stent are placed in healthy areas, and preferably avoid the stent overlapping any confluences. To do this, physicians typically use IVUS images generated during the pullback procedure. Physicians may also use additional IVUS images obtained by revisiting possible sites of interest, additional X-ray images generated simultaneously with the IVUS images, and information from the initial venogram. However, this process can be cumbersome due to the complexity of analyzing the IVUS images and correlating information from different imaging systems.

[0004] Upon designating a landing zone for a stent, the physician typically measures the length of the landing zone to designate the length of the stent for insertion into the lumen. In vascular procedures, the length of the landing zone is typically designated by positioning the IVUS imaging catheter such that fiducial markers located along the shaft of the IVUS imaging catheter overlap the intended landing zone. An additional clinician supporting the physician then manually counts the number of fiducial markers in the landing zone on the x-ray image. The physician then designates the stent length based on the count. However, this procedure is inefficient and prone to error due to the need for an additional clinician, the need to position the IVUS imaging catheter such that its fiducial markers overlap the landing zone, and the need to manually count the fiducial markers to determine their length. The fiducial markers are typically placed at coarse intervals of one centimeter along the shaft of the IVUS imaging catheter, and there may be 15 or more markers to count. Summary of the Invention [Problem to be solved by the invention]

[0005] Similar challenges arise when specifying stents for insertion into other lumens within the anatomy, including the coronary arteries, bile ducts, bronchial, and urinary lumens. However, a solution to one or more of these problems remains needed. [Means for solving the problem]

[0006] According to one aspect of the present disclosure, a system for determining the suitability of a location along a lumen as an end location for a stent is provided. The system includes: receiving a time sequence of endoluminal images representative of positions along the lumen, including at least a first endoluminal image, from an endoluminal imaging device; Identifying whether the first intraluminal image represents a suitable end point location for the stent; The system includes one or more processors configured to:

[0007] Further aspects, features, and advantages of the present disclosure will become apparent from the following detailed description of exemplary embodiments which proceeds with reference to the accompanying drawings. [Brief description of the drawings]

[0008] [Figure 1] FIG. 1 is a schematic diagram illustrating an example of a system 100 for determining the suitability of a location along a lumen as an end location of a stent, according to some aspects of the present disclosure. [Diagram 2] 1 is a flow chart illustrating an example of a method for determining suitability of a location along a lumen as an endpoint location for a stent, according to some aspects of the present disclosure. [Diagram 3] 1A-1C are schematic diagrams illustrating an example of determining suitability of a location along a lumen as an end location of a stent, according to some aspects of the present disclosure. [Figure 4] FIG. 15 is a schematic diagram illustrating an example of determining suitability of a position along a lumen as an end-point position for a stent, including step S150 of analyzing a temporal sequence of pre-treatment endoluminal images 1501 ..k, according to some aspects of the present disclosure. [Diagram 5] 1 is a schematic diagram illustrating a first X-ray image 180 and a second X-ray image 190 generated simultaneously with a first endoluminal image 140i and a second endoluminal image 140m, respectively, according to some embodiments of the present disclosure. [Figure 6] An endoluminal IVUS image 140u of an unhealthy, or "compacted," lumen in which the lumen area is significantly reduced compared to the average lumen area of ​​a similar patient cohort is shown. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] Examples of the present disclosure are provided with reference to the following description and drawings. In this description, for purposes of explanation, many specific details of particular examples are described. In this specification, reference to an "example", "embodiment", or similar language means that a feature, structure, or characteristic described in connection with an example is included in at least that one example. It should also be understood that features described in connection with one example may also be used in another example, and not all features are necessarily repeated in each example for the sake of brevity. For example, in this specification, reference is made to a system including one or more processors configured to perform one or more methods. It should be understood that one or more methods may alternatively be provided as a computer-implemented method, or may be implemented in a computer program product, or may be provided on a computer-readable storage medium, in a corresponding manner.

[0010] In the following description, reference is made to a system including one or more processors configured to perform various functions, which may be steps of a method including determining the suitability of a position along a lumen as an end point location for a stent. Reference is made to a lumen in the form of a vein within the vasculature. However, it should be understood that the method described in this disclosure may be used to determine the suitability of a position along other lumens within the vasculature as an end point location for a stent, such as an artery. Furthermore, it should be understood that the method may be used to determine the suitability of a position along a lumen within an anatomical structure in general as an end point location for a stent, including lumens within coronary arteries, bile ducts, bronchi, urinary tracts, and other regions of the anatomical structure.

[0011] In the methods described herein, reference is made to an endoluminal imaging device in the form of an IVUS imager, which may be provided in the form of, for example, an IVUS imaging catheter, an IVUS imaging guidewire, etc. However, it should be understood that an IVUS imager serves only as one example of an endoluminal imaging device, and the methods may be used with other types of endoluminal imaging devices, including OCT imagers, TEE probes, etc., as appropriate.

[0012] In the methods described herein, reference is also made to detection of the endoluminal imaging device in the x-ray image. In this regard, it should be understood that the x-ray image may be generated by various types of x-ray imaging systems. For example, the x-ray image may be generated by a planar imaging system that generates a planar x-ray image, or a volumetric x-ray imaging system that generates a volumetric x-ray image. A planar x-ray imaging system typically includes a support arm, such as a so-called "C-arm" or "O-arm," that supports the x-ray source-detector configuration. A planar x-ray imaging system may alternatively include a support arm having a different shape than these examples. A planar x-ray imaging system typically generates a planar x-ray image with the support arm held in a stationary position relative to the imaging region during acquisition of the image data. In contrast, a volumetric x-ray imaging system typically generates image data while rotating or stepping the x-ray source-detector configuration around the imaging region, and then reconstructs the image data obtained from multiple rotation angles into volumetric image data. Examples of volumetric x-ray imaging systems include computed tomography "CT" imaging systems, cone beam computed tomography "CBCT" imaging systems, and spectral CT imaging systems.

[0013] It should be noted that the methods disclosed herein may be provided as a non-transitory computer-readable storage medium including stored computer-readable instructions that, when executed by at least one processor, cause the at least one processor to perform the method. In other words, the computer-implemented method may be implemented in a computer program product. The computer program product may be provided by dedicated hardware, or by hardware capable of executing software in association with appropriate software. When provided by a processor, the functionality of the features of the method may be provided by a single dedicated processor, or by a single shared processor, or by multiple individual processors, some of which may be shared. The explicit use of the terms "processor" or "controller" should not be construed as exclusively referring to hardware capable of executing software, and may implicitly include, but is not limited to, digital signal processor "DSP" hardware, read only memory "ROM" for storing software, random access memory "RAM", non-volatile storage, and the like. Furthermore, examples of the present disclosure may take the form of a computer usable storage medium, or a computer program product accessible from a computer readable storage medium, the computer program product providing program code for use by or in association with a computer or any instruction execution system. For purposes of this description, a computer usable storage medium or computer readable storage medium can be any device that can have, store, communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. The medium can be an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system or device or propagation medium. Examples of computer readable media include semiconductor or solid state memories, magnetic tape, removable computer diskettes, random access memory "RAM", read only memory "ROM", rigid magnetic disks, and optical disks.Current examples of optical disks include Compact Disk-Read Only Memory "CD-ROM", Compact Disk-Read / Write "CD-R / W", Blu-Ray, and DVD.

[0014] As mentioned above, in the medical field, it is often necessary to determine the appropriate location, or "landing zone," of a stent within a lumen. To address this problem, the inventors have provided a system and method for determining the suitability of a location along a lumen as an end location for a stent. FIG. 1 is a schematic diagram illustrating an example of a system 100 for determining the suitability of a location along a lumen as an end location for a stent, according to some embodiments of the present disclosure. The system 100 includes: At least a first intraluminal image 140 is acquired from the intraluminal imaging device 130. i⊂1..n endoluminal image 140 representing a position along lumen 110, 1..n receiving a time sequence of (S110); First intraluminal image 140 i⊂1..n represents a suitable end location for the stent (S120); The device includes one or more processors 120 configured to:

[0015] As shown in Fig. 1, the system 100 may optionally include one or more of an intraluminal imaging device 130, an extracorporeal imaging system 240, such as an X-ray imaging system shown in Fig. 1, or alternatively another imaging system, one or more displays 250, and a patient bed 260. The system 100 may optionally include a user interface device, such as a keyboard, and / or a pointing device, such as a mouse (not shown in Fig. 1), for controlling execution of instructions of the method. The elements of the system are in communication with each other, as indicated by the interconnecting arrow lines in Fig. 1. By way of example, the intravascular imaging catheter may be an IVUS imaging catheter, such as the Visions PV .014P RX, available from Philips Healthcare of Best, The Netherlands.

[0016]

[0023] Figure 2 is a flow chart illustrating an example of a method for determining suitability of a location along a lumen as an end location for a stent, according to some aspects of the present disclosure. The operations illustrated in Figure 2 may be performed by one or more processors 120 of the system 100 illustrated in Figure 1. The flow chart illustrated in Figure 2 includes the receive S110 and identify S120 operations described above in connection with Figure 1, and may optionally include one or more of operations S130, S140, S150, and S160, as described in more detail below.

[0017] 3 is a schematic diagram illustrating an example of determining suitability of a location along a lumen as an end location for a stent, according to some aspects of the present disclosure. FIG. 3 illustrates the operations of receiving S110 and identifying S120 described above, and may optionally include operation S130, which is described in more detail below.

[0018] 1 to 3, in operation S110, an intraluminal image 140 1..n The temporal sequence of images 140 may be received via any form of data communication, including wired and wireless communication. As some examples, when wired communication is used, the communication may be via electrical or optical cables, and when wireless communication is used, the communication may be via RF or infrared signals, for example. 1..n may be received directly from the endoluminal imaging device 130 or indirectly, for example, via a computer-readable storage medium. 1..n may represent live images generated instantaneously before being received, or may be generated seconds, minutes, hours, or days, or longer, in advance. 1..n may be generated by an IVUS imager or another endoluminal imaging device, as described above. Each image in the temporal sequence of images may be referred to as an image frame.

[0019] Intraluminal images 140 1..n Before the operation of identification S120, the first intraluminal image 140 i⊂1..nWhether or not represents a suitable end position for the stent may be stored, for example, in memory MEM as shown in FIG. 3, or in a buffer (not shown in FIG. 3). Operation S120 may include displaying the associated text, or an icon, or another identifying information on a display device, or generating the identifying information via another visual or audio alert.

[0020] Various additional operations may be performed in combination with operations S110 and S120 described above.

[0021] Generally, the first intraluminal image 140 i⊂1..n The act of identifying S120 whether represents a suitable end location for the stent may be based on measurements of the lumen and / or the shape of the lumen, or a classification performed by an artificial intelligence algorithm.

[0022] As mentioned above, typically, the physician determines the landing zone of the stent such that the stent overlaps the diseased area and the end of the stent is placed in a healthy area, but preferably avoids the stent overlapping any confluence. Endoluminal images representing healthy areas typically have a luminal area close to the average luminal area of ​​the patient cohort. In contrast, endoluminal images representing diseased areas often contain compressions or thrombi whose luminal area is significantly smaller than the average luminal area of ​​the patient cohort. Confluences occur when there is a branching from a lumen to an adjacent lumen. At a confluence, the lumen may branch or the two lumens may merge into a single lumen. Endoluminal images representing a confluence typically have a luminal area significantly larger than the average luminal area of ​​the patient cohort. It can be difficult to place a stent at a confluence without the risk of the stent migrating over time. Thus, lumen measurements can be used to distinguish between healthy regions, diseased regions, and confluences, thereby determining whether an endoluminal image represents the proper end-point location for a stent.

[0023] The shape of the lumen in the endoluminal image may be similarly used to distinguish between healthy regions, unhealthy regions, and confluences, thereby determining whether the endoluminal image represents a suitable end location for the stent. For example, healthy regions of the lumen may have a shape close to circular, while diseased regions, such as compressions and thrombi, may have an irregular, i.e., non-circular, shape. The presence of irregular shapes, such as ovals, or sharp angles in the lumen often indicates diseased regions of the lumen. However, an irregular shape alone may be insufficient to diagnose a lumen as unhealthy, especially if the lumen area is not significantly lower than the average lumen area of ​​the patient cohort. Thus, in some instances, the distinction between healthy and unhealthy regions may be made based on a combination of lumen shape and lumen measurement, factors. At confluences, the shape of a healthy lumen may be circular within a major range of rotation angles around the lumen axis, and the shape of the lumen may bulge outward within a small range of rotation angles around the lumen axis, where the lumen wall is defined by the wall of the adjacent lumen.

[0024] Lumen area measurements and shapes may be obtained by segmenting the endoluminal image. Various image segmentation techniques are known for this purpose. Lumen measurements may include, for example, determining the maximum lumen width in the segmented endoluminal image and the corresponding width in an orthogonal direction. Other lumen dimensions, such as the difference or ratio between measurements of the minor and major axes of a cross-section of the lumen, may also be calculated. Lumen measurements or shapes may be determined, for example, by fitting an ellipse-like shape to the lumen in the segmented endoluminal image.

[0025] Analysis of the endoluminal image to determine whether it represents a healthy or diseased region or a confluence and therefore a suitable endpoint location for the stent may include weighting of one or more of the above lumen measurements, and lumen shape, factors.

[0026] Image classification techniques using artificial intelligence algorithms may also be used to analyze endoluminal images and therefore determine whether they represent suitable end-point locations for the stent. The endoluminal images may be segmented prior to analysis by the artificial intelligence algorithm, or alternatively the endoluminal images may not be pre-segmented. As an example, a neural network may be trained to classify endoluminal images as representing healthy regions or diseased regions such as compression and thrombosis, or confluences. Alternatively, the neural network may be trained to directly classify endoluminal images as representing suitable end-point locations for a stent or not. The neural network may be trained by inputting training data into the neural network, including images labeled with suitability, or one of the aforementioned classifications, and a more specific "ground truth" classification, and optimizing the parameters of the neural network based on the value of a loss function that represents the difference between the classification predicted by the neural network and the label.

[0027] Thus, the method shown in FIG. 2 may optionally follow the left-hand branch shown in FIG. Intraluminal images 140 1..n and comparing measurements of the lumen 110 represented in the segmented endoluminal image to expected measurements, and / or comparing the shape of the lumen 110 represented in the segmented endoluminal image to an expected shape, and / or Using artificial intelligence algorithms to image the endoluminal 1..n Classify the Here, the identification S120 is performed based on the analysis S130. By this, intraluminal image 140 1..n The method may include an act of analyzing a temporal sequence of S130.

[0028] Because the analysis operation S130 is performed automatically in the manner described above, improved workflow and / or consistent analysis is provided by the method.

[0029] In an alternative embodiment, the method follows the right branch shown in Fig. 2. This embodiment is illustrated with reference to Fig. 4, which is a schematic diagram illustrating an example of determining the suitability of a location along a lumen as an end point location for a stent, based on a pre-treatment endoluminal image 150, according to some aspects of the present disclosure. 1..k In this embodiment, the temporal sequence of endoluminal images 140 received and identified in operations S110 and S120, respectively, described above is analyzed S150. 1..n represents the in-treatment image and further operations, i.e. A pre-treatment endoluminal image 150 representative of the lumen 110 is obtained from the endoluminal imaging device 130. 1..k an operation S140 of receiving a temporal sequence of The pre-treatment endoluminal image 150 is scanned to determine whether the pre-treatment endoluminal image represents a suitable endpoint location for the stent. 1..k an operation S150 of analyzing the time sequence of 150 matching pre-procedure endoluminal images j⊂1..k The first intraluminal image 140 during the procedure is i⊂1..n an operation S160 of comparing the pre-treatment endoluminal image 150 with the pre-treatment endoluminal image 150; j⊂1..k Based on the results of the analysis S150, the operation is executed by one or more processors.

[0030] In this embodiment, the analysis operation S150 is also performed automatically. 1..k However, image 140 during treatment 1..n For example, the pre-treatment intraluminal image 150 1..k may be generated in a pre-procedure IVUS pullback procedure, and the intra-procedure image 140 1..n may be generated during a subsequent IVUS imaging procedure where the physician revisits sites deemed interesting from the pre-treatment IVUS pullback procedure. Performing the analysis on the pre-treatment images in this manner reduces the processing burden of determining appropriate endpoint locations for the stent, as this processing operation can be performed in the period between generating the pre-treatment and in-treatment images.

[0031] Various metrics are matched to pre-procedure endoluminal images 150 j⊂1..k A first intraluminal image 140 is i⊂1..n 150 pre-treatment intraluminal images 1..k For example, the value of the least square error function may be calculated based on the difference between the image intensities after the images are aligned with each other. Alternatively, a learned similarity metric may be used. A temporal constraint may also be applied to ensure that matching images in the sequence of pre-treatment and in-treatment images are coherent in the direction of pullback.

[0032] The pre-treatment endoluminal image 150 is scanned to determine whether the pre-treatment endoluminal image represents a suitable endpoint location for the stent. 1..k The act of analyzing the temporal sequence of the pre-treatment endoluminal image 150 may be performed in a manner similar to that described above in connection with the analysis performed in act S130. 1..k Analyzing the time sequence of S150 Pre-treatment intraluminal image 150 1..k and comparing measurements of the lumen 110 represented in the segmented pre-treatment endoluminal image with expected measurements and / or comparing the shape of the lumen represented in the segmented pre-treatment endoluminal image with an expected shape; and / or 150 pre-procedure endoluminal images using artificial intelligence algorithms 1..k and classifying the

[0033] As mentioned above, the analysis performed in operation S150 includes: Pre-treatment intraluminal image 150 1..k represents at least one of a healthy region of the lumen 110, a confluence in the lumen 110, a compression in the lumen 110, and a thrombosis in the lumen 110; may include:

[0034] In one example, the first endoluminal image 140 performed in operation S160. i⊂1..n and 150 pre-procedure intraluminal images 1..k The comparison between the pre-treatment endoluminal image 150 and the adjacent image 150 may be improved by basing the comparison between the image under consideration and its adjacent image. j⊂1..k The first intraluminal image 140 during the procedure is i⊂1..n 150 pre-treatment intraluminal images 1..k The operation S160 to be compared is The first intraluminal image 140 is scanned to determine if a match exists. i⊂1..n 150. 1..k comparing the image to one or more corresponding adjacent images from It further has:

[0035] This has the effect of enforcing temporal consistency between the in-treatment image stream and the pre-treatment images.

[0036] In one example, a pre-procedure endoluminal image 150 1..k The results of analyzing (S150) the temporal sequence of the endoluminal images may be displayed. This allows the physician to determine which locations along the lumen may be worth revisiting with an endoluminal imager to confirm a suitable location for the stent. The results of the analysis may be displayed in a variety of ways. For example, the results may be displayed as a pre-treatment endoluminal image 150 along the axis of the lumen. 1..k 10 and by including markers or color coding indicating the suitability of each location along the lumen as an end point location of the stent, or vice versa. i⊂1..n An x-ray image including positions along the lumen 110 represented by may be displayed.

[0037] In another example, endoluminal images are generated at different locations along the lumen and the locations of the endoluminal images are indicated in a common X-ray image. Anatomical information provided by the common X-ray image may be useful in identifying the location as suitable for the stent. In this example, the X-ray images are generated simultaneously with the endoluminal images and a mapping operation is performed in which the location of one of the endoluminal images along the lumen is mapped from one of the X-ray images to a corresponding location in another X-ray image.

[0038] This example illustrates a first intraluminal image 140, respectively, according to some aspects of the present disclosure. i and a second intraluminal image 140 m The procedure will be described with reference to FIG. 5, which is a schematic diagram showing a first X-ray image 180 and a second X-ray image 190 generated simultaneously. The X-ray images 180, 190 shown in FIG. 5 include an intraluminal imaging catheter at respective first and second positions 160, 170 along the vessel lumen 110. The X-ray images 180, 190 may include dense structures such as bones that serve as landmarks (not shown in FIG. 5). However, the vessel lumen 110 is mostly invisible in the X-ray images due to the small difference in X-ray attenuation values ​​between the luminal tissue and water. Thus, the vessel lumen 110 is shown in FIG. 5 with a dashed outline. The first and second positions 160, 170 may correspond to positions of the intraluminal imaging catheter along the lumen during an intraluminal imaging catheter pull-back procedure. The positions may alternatively correspond to positions of the intraluminal imaging catheter where the physician guided the intraluminal imaging catheter into the lumen and where the physician considers it suitable to place the end of the stent. The mapping operation includes obtaining a second endoluminal image 140 from the lumen in a second x-ray image 190. m 5 by dashed horizontal lines indicating the mapping of location 170 to a corresponding location in the first x-ray image 180.

[0039] In the example shown in FIG. i and 140 m Both represent IVUS images of healthy regions of the lumen. iA healthy lumen, represented in endoluminal image 140, has a circular shape, as represented by the dark central area in the image. m The lumen depicted in is still considered healthy because its area is not significantly reduced compared to the average lumen area of ​​a similar patient cohort. In comparison, FIG. 6 shows an endoluminal IVUS image 140 of an unhealthy or "compressed" lumen. u , with the luminal area being significantly reduced compared to the mean luminal area of ​​a similar patient cohort.

[0040] In the example shown in FIG. 5, in addition to the receiving and identifying operations S110 and S120 described above with reference to FIG. 2, a first intraluminal image 140 i⊂1..n represents a first location 160 along the lumen 110, and the endoluminal image 140 1..n The operation S110 of receiving the temporal sequence of images S110 generates a second endoluminal image 140 representing a second location 170 along the lumen 110. m⊂1..n The method also includes receiving Receiving a first X-ray image 180 and a second X-ray image 190, the first X-ray image and the second X-ray image being respectively a first endoluminal image 140; i⊂1..n and a second intraluminal image 140 m⊂1..n and a step, First and second endoluminal images 140 in respective first and second X-ray images 180, 190 i⊂1..n , 140 m⊂1..n determining the positions 160, 170 of First and second endoluminal images 140 from respective first and second X-ray images 180, 190 i⊂1..n , 140 m⊂1..n mapping one of the determined positions 160, 170 onto the other of the first and second X-ray images 180, 190; Includes.

[0041] In this example, the X-ray images 180, 190 are the same as the endoluminal image 140. 1..nThe X-ray images 180, 190 may be received by any means of data communication, as described above with respect to the temporal sequence of the X-ray images 180, 190. ... 1..n Alternatively, contemporaneous x-ray and endoluminal images may be provided by selecting an x-ray or endoluminal image from a stream of images, providing a contemporaneous image that is generated closest in time to the respective endoluminal or x-ray image. In one example, the first and second endoluminal images 140 i⊂1..n , 140 m⊂1..n , in response to the generation of the first and second X-ray images 180, 190, respectively, 1..n The time sequence is automatically selected.

[0042] Various techniques can be used to extract the first and second endoluminal images 140 from the respective first and second X-ray images 180, 190. i⊂1..n , 140 m⊂1..n The determined positions 160, 170 may be used in mapping one of the determined positions 160, 170 to the other of the first and second x-ray images 180, 190. In one example, the first and second x-ray images 180, 190 may be overlaid to map positions from one x-ray image to the other x-ray image. In another example, the mapping may be performed using a performing an image-based registration between the first X-ray image 180 and the second X-ray image 190; Calculating a spatial transformation between corresponding image features in the first and second X-ray images 180, 190 based on the image-based registration; Including, Here, the mapping is performed based on the calculated spatial transformation.

[0043] Performing the image-based registration described above advantageously compensates for any motion between the generation of the first and second X-ray images 180, 190, thereby improving the accuracy of the mapped positions.

[0044] In general, the location 160, 170 of the endoluminal image along the lumen may be determined based on detection of the imaging portion 220 or another portion of the endoluminal imaging device in the x-ray image 180, 190. The imaging portion 220 typically comprises a radiopaque material that is visible in the x-ray image. By detecting the imaging portion 220, the relative location of the imaging region of the lumen imaged in the endoluminal image may be determined based on the known field of view of the endoluminal imaging device. For example, some IVUS imaging catheters typically have a field of view that is circumferential about the longitudinal axis of the IVUS imaging catheter and normal to its imaging portion 220. Other IVUS imaging catheters have a field of view that is circumferential about the longitudinal axis of the IVUS imaging catheter and also tilted in a forward looking direction relative to the imaging portion 220. Thus, the location along the lumen corresponding to the endoluminal image may be determined based on knowledge of the location of the detected imaging portion and the known imaging field of view.

[0045] One or more fiducial markers may be disposed on the endoluminal imaging device. The fiducial markers include radiopaque material that can be detected in an x-ray image. Thus, the relative position of the imaged region of the lumen can be determined based on the detected positions of the radiopaque markers in a similar manner.

[0046] Various techniques may be used in determining the location 160, 170 of the endoluminal image in each x-ray image 180, 190. For example, the use of computer vision object detection techniques, image segmentation techniques, and artificial intelligence algorithms are contemplated. As an example, when an artificial intelligence algorithm is used, the artificial intelligence algorithm may be provided by a neural network trained to identify the location of the imaging portion 220 of the endoluminal imaging catheter in the x-ray image. The neural network may be trained to identify the location by inputting x-ray image training data including the ground truth location of the imaging portion 220 to the neural network and adjusting the parameters of the neural network using a loss function that represents the difference between the predicted location of the imaging portion 220 predicted by the neural network and its ground truth location from the x-ray image training data. The ground truth location may be identified in the x-ray image training data by an expert. Suitable types of neural networks for this purpose include U-Net, UNet++, VNet, convolutional neural network "CNN", and region-based CNN "RCNN", among others.

[0047] Thus, the first and second intraluminal images 140 in the respective first and second X-ray images 180, 190 i⊂1..n , 140 m⊂1..n The action of determining the position of performing computer vision object detection techniques; and / or Segmenting the first and second X-ray images 180, 190; and / or applying an artificial intelligence algorithm to each of the first and second x-ray images, the artificial intelligence algorithm being trained to detect the imaging portion of the endoluminal imaging device; This may include detecting the position of the imaging portion 220 of the endoluminal imaging device 130 in the respective first and second x-ray images 180, 190 by

[0048] The methods or operations performed by the one or more processors may optionally include offsetting the detected position 160, 170 of the imaging portion 220 along the lumen 110 by a predetermined distance that represents the difference in position between the imaging portion 220 of the endoluminal imaging device and the imaged area of ​​the lumen 110 imaged by the endoluminal imaging device 130. In doing so, the exact position of the endoluminal image may be determined in the x-ray image.

[0049] As described above, once the landing zone of the stent is determined, the physician typically measures the length of the landing zone to specify the length of the stent for insertion into the lumen. In one example, the method also includes: i⊂1..n , 140 m⊂1..n The method includes determining a length 210 of the portion of the lumen between the positions 160, 170 of the intraluminal imaging device 130. In this example, the intraluminal imaging device 130 includes a plurality of fiducial markers 200 positioned axially along the length of the device 130. 1..y The mapping includes: the X-ray images 180, 190 onto which the locations 160, 170 are mapped are aligned with the first and second endoluminal images 140 i⊂1..n , 140 m⊂1..n One or more fiducial markers 200 between the positions 160, 170 1..y The method is performed to include the steps of: First and second intraluminal images 140 i⊂1..n , 140 m⊂1..n The length 210 of the portion of the lumen 110 between the locations 160, 170 of the first and second endoluminal images 140 in the x-ray images 180, 190 on which the locations are mapped is i⊂1..n , 140 m⊂1..n calculating based on a count of the number of fiducial markers between the positions 160, 170; Includes.

[0050] An example of this is described with reference to FIG. 5. Fiducial markers 200 positioned axially along the length of the intraluminal imaging device. 1..yThe fiducial markers 210 may comprise a radiopaque material, such as stainless steel, gold, platinum, etc., and thus are detectable in the x-ray images 180, 190. Advantageously, in this example, the fiducial markers are inherently present in one of the x-ray images, and therefore, there is no need to reposition the endoluminal imaging device in the x-ray image to perform the length measurement. Furthermore, the location of the fiducial markers used for the length measurement is essentially aligned with one end of the portion of the lumen to be measured, simplifying the measurement. The calculated length 210 of said portion of the lumen 110 may correspond to a recommended length of a stent for insertion into the lumen 110.

[0051] The fiducial markers may be detected in the X-ray image using image analysis techniques. Such techniques are known from the field of computer vision. Path calculation methods such as forward propagation or fast marching may be used to determine the order of the fiducial markers. Since the counting is performed automatically in the present method, no manual counting by the physician is required, thus easing the workflow and / or avoiding counting errors. Manual counting of fiducial markers is prone to errors due to the long length of the diseased lumen treated in deep vein intervention procedures. In such procedures, the length of the diseased lumen may exceed 15 centimeters or more.

[0052] In practice, the measured length 210 of the lumen 110 may not correspond to an integer number of fiducial markers. In this situation, interpolation may be used to more accurately determine the length 210 of the portion of the lumen 110. Thus, in some examples, the act of calculating the length 210 of the portion of the lumen 110 may be Reference Marker 200 1..y and the first location 160 and / or the second location 170; may include:

[0053] In one example, the intraluminal image 140 shown in FIG. 1..nrepresents an intraluminal image of the lumen, and the method also includes receiving a pre-treatment image of the lumen. In this example, similar to that described above in connection with FIG. 4, the pre-treatment image is analyzed to determine whether an image in the pre-treatment image represents a suitable endpoint location for the stent, and determining the suitability of the intraluminal image as an endpoint location for the stent includes comparing the intraluminal image to the pre-treatment image. Referring to FIG. 5, in this example, endoluminal image 140 is analyzed to determine whether an image in the pre-treatment image represents a suitable endpoint location for the stent. 1..n represents an in-treatment image, and the method further comprises: A pre-treatment endoluminal image 150 representative of the lumen 110 is obtained from the endoluminal imaging device 130. 1..k and a step S140 of receiving a time sequence of The pre-treatment endoluminal image 150 is scanned to determine whether the pre-treatment endoluminal image represents a suitable endpoint location for the stent. 1..k A step S150 of analyzing To identify respective first and second matching pre-treatment endoluminal images, the first and second during-treatment endoluminal images 140 are i⊂1..n , 140 m⊂1..n 150 pre-treatment intraluminal images 1..k and comparing having The identifying step S120 includes identifying the first and second in-procedure endoluminal images 140. i⊂1..n , 140 m⊂1..n based on the results of the analysis S150 of the first and second matching pre-treatment endoluminal images, identifying whether the first and second matching pre-treatment endoluminal images represent a suitable end point location for the stent, and displaying the results 230 of the analysis S150 of the first and / or second matching pre-treatment endoluminal images.

[0054] The results 230 of the analysis S150 may be displayed by providing text labels on the display, for example, with exemplary health and compaction labels as shown in Figures 5 and 6, or may be displayed in another manner. For example, the results may be displayed with icons or by highlighting locations in the x-ray images 180, 190 with a color or intensity indicating suitability as a stent location and as an endpoint location. In another example, markers may be provided in the x-ray image 180 that includes the mapped locations. The markers may highlight areas of the x-ray image with a color or intensity, or a line or symbol indicating the result of the analysis, i.e., "healthy," or "confluence," or "compaction," or "thrombosis."

[0055] As discussed above, performing analysis on pre-treatment images in this manner reduces the processing burden of determining appropriate endpoint locations for the stent, since this processing can be performed in the time interval between the generation of the pre-treatment and in-treatment images.

[0056] This example may also include the act of receiving user input identifying the locations 160, 170 along the lumen 110 as representing suitable endpoint locations for the stent. This provides the physician with an opportunity to confirm suitability and / or select an alternate location as an endpoint location for the stent. According to this example, the method further includes: receiving user input identifying first and second locations 160, 170 along the lumen 110 as representing suitable end locations for the stent; having Here, the step of calculating the length 210 of the portion of the lumen 110 is performed automatically, subject to receiving user input identifying the first and second positions 160, 170 along the lumen 110 as representing suitable end positions for the stent.

[0057] Any of the above methods may First and / or second intraluminal image 140 i⊂1..n , 140 m⊂1..n , Pre-treatment intraluminal image 1501..k Analysis of the time sequence of 230 Analysis of the matched pre-procedure endoluminal images revealed that Upon analysis of the first and / or second matching pre-treatment endoluminal images, First and / or second X-ray images 180, 190; a calculated length 120 of the portion of the lumen 110; The method may include displaying one or more of:

[0058] The above methods may be incorporated into clinical procedures in different ways, as illustrated by the above examples. Note that each of these examples may be preceded by an initial X-ray imaging procedure using contrast agents to provide a map of the vasculature. This initial procedure, e.g., venography, may include the use of Digital Subtraction Angiography (DSA) techniques.

[0059] The above examples should be understood as illustrative and not limiting of the present disclosure. Further examples are also contemplated. For example, the exemplary method described with respect to the system 100 may be provided in a corresponding manner as a computer-implemented method, a computer program product, or by a computer-readable storage medium. It should be understood that the features described with respect to any one example may be used alone or in combination with other described features, and may be used in combination with one or more other features of the example, or with combinations of other examples. Moreover, equivalents and modifications not described above may also be used without departing from the scope of the present invention, as defined in the appended claims.

[0060] In the claims, the word "comprise" does not exclude other elements or operations, and the indefinite articles "a" or "an" do not exclude a plurality. The mere fact that certain features are recited in mutually different dependent claims does not indicate that a combination of these features cannot be used to advantage. Any reference signs in the claims should not be interpreted as limiting their scope.

Claims

1. 1. A system for determining suitability of a location along a lumen for an end location of a stent, comprising: receiving a temporal sequence of endoluminal images representative of positions along the lumen, the temporal sequence including at least a first endoluminal image, from an endoluminal imaging device; identifying whether the first intraluminal image represents a suitable end point location for a stent; 23. A system comprising at least one processor configured to:

2. The at least one processor Segmenting the endoluminal image and comparing measurements of the lumen represented in the segmented endoluminal image with expected measurements, and / or comparing a shape of the lumen represented in the segmented endoluminal image with an expected shape, and / or classifying the endoluminal images using an artificial intelligence algorithm; and further configured to analyze the temporal sequence of endoluminal images by: the identification is based on the analysis. The system of claim 1 .

3. The endoluminal image represents an intra-procedure image, and the at least one processor further comprises: receiving a temporal sequence of pre-treatment endoluminal images representative of the lumen from the endoluminal imaging device; analyzing the temporal sequence of pre-treatment endoluminal images to determine whether the pre-treatment endoluminal images represent a suitable endpoint location for the stent; comparing the first during-treatment endoluminal image to the pre-treatment endoluminal image to identify a matching pre-treatment endoluminal image; It is configured as follows: the identification is based on a result of the analysis of the matching pre-treatment endoluminal image. The system of claim 1 .

4. Analyzing the temporal sequence of pre-treatment endoluminal images includes: Segmenting the pre-treatment endoluminal image and comparing measurements of the lumen represented in the segmented pre-treatment endoluminal image with expected measurements, and / or comparing a shape of the lumen represented in the segmented pre-treatment endoluminal image with an expected shape, and / or classifying said pre-treatment endoluminal images using an artificial intelligence algorithm; The apparatus of claim 3 , comprising:

5. The analysis determining whether the pre-treatment endoluminal image represents at least one of a healthy region of the lumen, a confluence in the lumen, a compression in the lumen, and a thrombosis in the lumen; The system of claim 3 , comprising:

6. comparing the first pre-treatment endoluminal image to the pre-treatment endoluminal image to identify the matching pre-treatment endoluminal image; comparing one or more adjacent endoluminal images of the first during-treatment endoluminal image with one or more corresponding adjacent images from the pre-treatment endoluminal image to determine if a match exists; The system of claim 3 further comprising:

7. The at least one processor further comprises: outputting the results of analyzing the temporal sequence of raw endoluminal images on a display; and / or outputting to the display an x-ray image including the location along the lumen represented by the first endoluminal image; The system of claim 3 , configured to:

8. The first endoluminal image represents a first location along the lumen, and receiving the temporal sequence of endoluminal images includes receiving a second endoluminal image representing a second location along the lumen, and the at least one processor further comprises: receiving a first X-ray image and a second X-ray image, the first X-ray image and the second X-ray image being generated simultaneously with the first endoluminal image and the second endoluminal image; determining a position of the first and second endoluminal images in the first and second x-ray images, respectively; mapping one of the determined locations of the first and second endoluminal images from the respective first or second X-ray images onto the other of the first and second X-ray images. The system of claim 1 configured to:

9. the endoluminal imaging device further comprises a plurality of fiducial markers disposed axially along a length of the device, and the mapping is performed such that the x-ray image onto which the locations are mapped includes one or more of the fiducial markers between the locations of the first and second endoluminal images, and the at least one processor further comprises: calculating a length of the portion of the lumen between the locations of the first and second endoluminal images based on a count of the number of the fiducial markers between the locations of the first and second endoluminal images in the X-ray image to which the locations are mapped; The system of claim 8 , configured to:

10. The mapping may include: performing an image-based registration between the first and second x-ray images; calculating a spatial transformation between corresponding image features in the first and second x-ray images based on the image-based registration; Including, the mapping is performed based on the calculated spatial transformation. The system of claim 8.

11. Determining the positions of the first and second endoluminal images in the first and second x-ray images, respectively, comprises: performing computer vision object detection techniques; and / or Segmenting the first and second X-ray images; and / or applying an artificial intelligence algorithm to each of the first and second x-ray images, the artificial intelligence algorithm being trained to detect the imaging portion of the endoluminal imaging device; and Optionally, offsetting the detected position of the imaging portion along the lumen by a predetermined distance representing a difference in position between the imaging portion of the endoluminal imaging device and an imaging area of ​​the lumen imaged by the endoluminal imaging device; 9. The system of claim 8, further comprising detecting a position of an imaging portion of the endoluminal imaging device in each of the first and second x-ray images by:

12. The system of claim 8 , wherein the first and second endoluminal images are automatically selected from the temporal sequence of endoluminal images responsive to generation of the first and second x-ray images, respectively.

13. The endoluminal image represents an intra-procedure image, and the at least one processor further comprises: receiving a temporal sequence of pre-treatment endoluminal images representative of the lumen from the endoluminal imaging device; analyzing the temporal sequence of pre-treatment endoluminal images to determine whether the pre-treatment endoluminal images represent a suitable endpoint location for the stent; comparing the first and second during-treatment endoluminal images to the pre-treatment endoluminal image to identify respective first and second matching pre-treatment endoluminal images; It is configured as follows: the identifying includes identifying whether the first and second in-procedure endoluminal images represent suitable endpoint locations for the stent based on results of the analysis of the first and second matching pre-procedure endoluminal images, and displaying results of the analysis of the first and / or second matching pre-procedure endoluminal images. The system of claim 8.

14. The at least one processor further comprises: receiving user input identifying first and second locations along the lumen as representing suitable endpoint locations for the stent; It is configured as follows: calculating the length of the portion of the lumen is performed automatically conditioned on receiving a user input identifying the first and second positions along the lumen as representing suitable end positions for the stent. The system of claim 13.

15. The at least one processor the first and / or second intraluminal images; Analysis of the time sequence of pre-treatment endoluminal images reveals: Analysis of the matching pre-treatment endoluminal images results in: As a result of analyzing the first and / or second matching pre-treatment endoluminal images, the first and / or second X-ray images, the calculated length of the portion of the lumen; The system of claim 13 , further configured to output at least one of the following to a display:

Images