Method, device, electronic device and storage medium for predicting stent implantation effect

Abstract

The present application provides a prediction method, device, electronic device and storage medium for the effect of stent implantation, the method comprising: obtaining an OCT image sequence and an IVUS image sequence for a diseased blood vessel; the OCT image sequence comprises an OCT image corresponding to each target section in the diseased blood vessel; the IVUS image sequence comprises an IVUS image corresponding to each target section; based on the OCT image and IVUS image corresponding to each target section, a lumen area mask, a calcification area mask, and a vascular external elastic membrane area mask are generated; the OCT image sequence, the IVUS image sequence, the lumen area mask, the calcification area mask, the vascular external elastic membrane area mask and the stent parameters of the target stent are input into a stent prediction model, and the implantation effect parameters of the diseased blood vessel and the expansion parameters of the target stent are predicted to represent the target stent implanted into the diseased blood vessel. The method is conducive to improving the accuracy of stent selection by physicians.

Description

Technical Field

[0001] The present application relates to the field of stent technology, and in particular to a method, device, electronic device and storage medium for predicting stent implantation effects. Background Art

[0002] Coronary heart disease is a major cardiovascular disease that threatens human health. Its common cause is thrombosis caused by ruptured atherosclerotic plaques or coronary artery stenosis, which leads to insufficient blood supply to the myocardium. Coronary intervention is one of the most commonly used methods to treat coronary heart disease. Physicians implant a suitable stent in the narrowed coronary lesion (i.e., in the diseased blood vessel), and then expand the stent with a balloon to restore the blocked blood vessel segment to its normal size, thereby restoring blood supply to the myocardium. However, it is not easy to accurately select a stent suitable for the patient's diseased blood vessel and ensure that the diseased blood vessel is fully expanded.

[0003] If the stent is not expanded enough, it will lead to poor stent adhesion and insufficient vascular expansion, which may further cause restenosis in the stent and stent thrombosis, directly endangering the patient's life and health; if the stent is expanded too much, it may cause edge dissection, tissue prolapse, etc., which may significantly increase the probability of secondary coronary heart disease. Therefore, it is very important to accurately select a stent suitable for the patient's diseased blood vessels.

[0004] In the prior art, when doctors select stents suitable for patients' diseased blood vessels, they can usually only select them based on their own subjective experience. Therefore, it is possible that the stent parameters of the selected stent are not suitable, resulting in poor expansion or poor adhesion or edge dissection or tissue prolapse caused by excessive expansion. Therefore, the prior art urgently needs a method that can provide doctors with direct medical reference when selecting stents suitable for patients, so that doctors can more accurately select stents suitable for patients. Summary of the invention

[0005] In view of this, the purpose of this application is to provide a prediction method, device, electronic device and storage medium for the effect of stent implantation, which can provide direct medical reference information for doctors when selecting stents, thereby helping to improve the accuracy of doctors' stent selection.

[0006] In a first aspect, the present application provides a method for predicting the effect of stent implantation, comprising:

[0007] Acquire an OCT image sequence and an IVUS image sequence for imaging within a diseased blood vessel; the OCT image sequence includes OCT images corresponding to each target section within the diseased blood vessel; the IVUS image sequence includes IVUS images corresponding to each target section;

[0008] Based on the OCT image and the IVUS image corresponding to each target section, a lumen region mask, a calcification region mask, and a vascular external elastic membrane region mask corresponding to the target section are generated;

[0009] The OCT image sequence, the IVUS image sequence, the lumen area mask, the calcification area mask, the vascular extraelastic membrane area mask and the stent parameters of the target stent are input into the trained stent prediction model to predict the implantation effect parameters of the diseased blood vessel after the target stent is implanted into the diseased blood vessel and the expansion parameters of the target stent.

[0010] In combination with the first aspect, the embodiment of the present application provides a first possible implementation of the first aspect, wherein the OCT image sequence and the IVUS image sequence are obtained by imaging the diseased blood vessel by the same intravascular imaging device during the same invasion; or, the OCT image sequence and the IVUS image sequence are obtained by imaging at different times;

[0011] When the OCT image sequence and the IVUS image sequence are obtained by imaging at different times, the OCT image sequence and the IVUS image sequence are registered in the following manner:

[0012] According to the target cross-section corresponding to each of the OCT images and the target cross-section corresponding to each of the IVUS images, the OCT image and the IVUS image corresponding to the same target cross-section are registered.

[0013] In combination with the first aspect, the embodiment of the present application provides a second possible implementation of the first aspect, wherein the step of generating a lumen area mask, a calcification area mask, and a vascular external elastic membrane area mask corresponding to each target section based on the OCT image and the IVUS image corresponding to the target section includes:

[0014] For each IVUS image corresponding to the target section, generating a mask of the vascular external elastic membrane region corresponding to the target section according to the vascular external elastic membrane region segmented from the IVUS image;

[0015] For each of the OCT images corresponding to the target cross section, generating the lumen region mask and the calcification region mask corresponding to the target cross section according to the lumen region and the calcification region segmented from the OCT image;

[0016] Alternatively, for the OCT image and the IVUS image corresponding to each target section, the lumen region mask and the calcification region mask corresponding to the target section are generated according to the lumen region and the calcification region segmented from the OCT image and the IVUS image.

[0017] In combination with the first aspect, the embodiment of the present application provides a third possible implementation of the first aspect, wherein, after generating the lumen area mask, the calcification area mask, and the vascular external elastic membrane area mask corresponding to each target section based on the OCT image and the IVUS image corresponding to the target section, the method further includes:

[0018] Determine a first lumen cross-sectional area in the proximal reference section and a second lumen cross-sectional area in the distal reference section according to the lumen region masks respectively corresponding to the proximal reference section and the distal reference section in the target section; wherein the proximal reference section and the distal reference section are respectively located at two ends of the diseased blood vessel;

[0019] The OCT image sequence, the IVUS image sequence, the first lumen cross-sectional area, the second lumen cross-sectional area, the position of the proximal reference section, the position of the distal reference section, and the lumen area mask corresponding to each target section are input into the trained stent selection model, and the stent parameters of at least one of the target stents are output.

[0020] In combination with the third possible implementation of the first aspect, the embodiment of the present application provides a fourth possible implementation of the first aspect, wherein the stent parameters include: stent cross-sectional size, stent length, stent placement position, expansion balloon size, and expansion pressure; the implantation effect parameters include: the third lumen cross-sectional area of ​​the diseased vessel on each target cross-section after expansion by the target stent, the probability of poor adhesion between the target stent and the diseased vessel, the probability of edge dissection of the diseased vessel, and the probability of tissue prolapse of the diseased vessel; the expansion parameter is the expansion rate.

[0021] In combination with the fourth possible implementation of the first aspect, the embodiment of the present application provides a fifth possible implementation of the first aspect, wherein the OCT image sequence, the IVUS image sequence, the lumen area mask, the calcification area mask, the vascular external elastic membrane area mask, and the stent parameters of the target stent are input into the trained stent prediction model to predict the implantation effect parameters of the diseased blood vessel after the target stent is implanted into the diseased blood vessel, including:

[0022] The OCT image sequence, the IVUS image sequence, the lumen region mask, the calcification region mask, and the vascular external elastic membrane region mask are respectively generated through the first encoder and the first decoder corresponding to each other to form the first feature matrix corresponding to each other;

[0023] Passing a second feature matrix obtained by fusing the first feature matrix through a second encoder to generate a third feature matrix;

[0024] Input the third characteristic matrix and the stent parameters into the fully connected layer, output the third lumen cross-sectional area of ​​the diseased blood vessel on each target cross-section after expansion by the target stent; and generate a fourth characteristic matrix by using the third characteristic matrix through a second decoder;

[0025] The fifth characteristic matrix obtained by fusing the fourth characteristic matrix, the pressure characteristic matrix and the stent cross-sectional size matrix is ​​input into the convolution layer, and the probability of poor adhesion between the target stent and the diseased blood vessel, the probability of edge dissection of the diseased blood vessel and the probability of tissue prolapse of the diseased blood vessel on each target section are output.

[0026] In combination with the fifth possible implementation of the first aspect, the embodiment of the present application provides a sixth possible implementation of the first aspect, wherein the predicting of the expansion parameter of the target stent after the target stent is implanted into the diseased blood vessel includes:

[0027] For each target cross-section, the ratio of the first area to the second area corresponding to the target cross-section is determined as the expansion rate of the target stent on the target cross-section; wherein the first area is the third lumen cross-sectional area, or the cross-sectional area of ​​the target stent after expansion; and the second area is any one of the first lumen cross-sectional area, the second lumen cross-sectional area, the average area of ​​the first lumen cross-sectional area and the second lumen cross-sectional area.

[0028] In a second aspect, the present application also provides a device for predicting stent implantation effects, comprising:

[0029] An acquisition module, used to acquire an OCT image sequence and an IVUS image sequence for imaging within a diseased blood vessel; the OCT image sequence includes OCT images corresponding to each target section within the diseased blood vessel; the IVUS image sequence includes IVUS images corresponding to each target section;

[0030] A generating module, for generating a lumen region mask, a calcification region mask, and a vascular external elastic membrane region mask corresponding to each target section based on the OCT image and the IVUS image corresponding to the target section;

[0031] The prediction module is used to input the OCT image sequence, the IVUS image sequence, the lumen area mask, the calcification area mask, the vascular extraelastic membrane area mask and the stent parameters of the target stent into the trained stent prediction model, and predict the implantation effect parameters of the diseased blood vessel after the target stent is implanted into the diseased blood vessel and the expansion parameters of the target stent.

[0032] In combination with the second aspect, the embodiment of the present application provides a first possible implementation of the second aspect, wherein the OCT image sequence and the IVUS image sequence are obtained by imaging the diseased blood vessel by the same intravascular imaging device during the same invasion; or, the OCT image sequence and the IVUS image sequence are obtained by imaging at different times;

[0033] When the OCT image sequence and the IVUS image sequence are obtained by imaging at different times, the OCT image sequence and the IVUS image sequence are registered in the following manner:

[0034] According to the target cross-section corresponding to each of the OCT images and the target cross-section corresponding to each of the IVUS images, the OCT image and the IVUS image corresponding to the same target cross-section are registered.

[0035] In combination with the second aspect, an embodiment of the present application provides a second possible implementation of the second aspect, wherein the generation module, when used to generate a lumen area mask, a calcification area mask, and a vascular external elastic membrane area mask corresponding to each target section based on the OCT image and the IVUS image corresponding to the target section, is specifically used to:

[0036] For each IVUS image corresponding to the target section, generating a mask of the vascular external elastic membrane region corresponding to the target section according to the vascular external elastic membrane region segmented from the IVUS image;

[0037] For each of the OCT images corresponding to the target cross section, generating the lumen region mask and the calcification region mask corresponding to the target cross section according to the lumen region and the calcification region segmented from the OCT image;

[0038] Alternatively, for the OCT image and the IVUS image corresponding to each target section, the lumen region mask and the calcification region mask corresponding to the target section are generated according to the lumen region and the calcification region segmented from the OCT image and the IVUS image.

[0039] In combination with the second aspect, the embodiment of the present application provides a third possible implementation of the second aspect, which further includes:

[0040] A determination module, configured to determine a first lumen cross-sectional area in the proximal reference section and a second lumen cross-sectional area in the distal reference section according to the lumen region masks corresponding to the proximal reference section and the distal reference section in the target section, after the generation module generates a lumen region mask, a calcification region mask, and a vascular external elastic membrane region mask corresponding to the target section based on the OCT image and the IVUS image corresponding to the target section; wherein the proximal reference section and the distal reference section are respectively located at two ends of the diseased blood vessel;

[0041] An input module is used to input the OCT image sequence, the IVUS image sequence, the first lumen cross-sectional area, the second lumen cross-sectional area, the position of the proximal reference section, the position of the distal reference section, and the lumen area mask corresponding to each target section into a trained stent selection model, and output the stent parameters of at least one of the target stents.

[0042] In combination with the third possible implementation of the second aspect, the embodiment of the present application provides a fourth possible implementation of the second aspect, wherein the stent parameters include: stent cross-sectional size, stent length, stent placement position, expansion balloon size, and expansion pressure; the implantation effect parameters include: the third lumen cross-sectional area of ​​the diseased vessel on each target cross-section after expansion by the target stent, the probability of poor adhesion between the target stent and the diseased vessel, the probability of edge dissection of the diseased vessel, and the probability of tissue prolapse of the diseased vessel; the expansion parameter is the expansion rate.

[0043] In combination with the fourth possible implementation of the second aspect, the embodiment of the present application provides a fifth possible implementation of the second aspect, wherein the prediction module is used to input the OCT image sequence, the IVUS image sequence, the lumen area mask, the calcification area mask, the vascular external elastic membrane area mask, and the stent parameters of the target stent into the trained stent prediction model, and predict the implantation effect parameters of the diseased blood vessel after the target stent is implanted into the diseased blood vessel and the expansion parameters of the target stent, specifically for:

[0044] The OCT image sequence, the IVUS image sequence, the lumen region mask, the calcification region mask, and the vascular external elastic membrane region mask are respectively generated through the first encoder and the first decoder corresponding to each other to form the first feature matrix corresponding to each other;

[0045] Passing a second feature matrix obtained by fusing the first feature matrix through a second encoder to generate a third feature matrix;

[0046] Input the third characteristic matrix and the stent parameters into the fully connected layer, output the third lumen cross-sectional area of ​​the diseased blood vessel on each target cross-section after expansion by the target stent; and generate a fourth characteristic matrix by using the third characteristic matrix through a second decoder;

[0047] The fifth characteristic matrix obtained by fusing the fourth characteristic matrix, the pressure characteristic matrix and the stent cross-sectional size matrix is ​​input into the convolution layer, and the probability of poor adhesion between the target stent and the diseased blood vessel, the probability of edge dissection of the diseased blood vessel and the probability of tissue prolapse of the diseased blood vessel on each target section are output.

[0048] In combination with the fifth possible implementation of the second aspect, the embodiment of the present application provides a sixth possible implementation of the second aspect, wherein the prediction module, when used to predict the expansion parameter of the target stent after the target stent is implanted into the diseased blood vessel, is specifically used to:

[0049] For each target cross-section, the ratio of the first area to the second area corresponding to the target cross-section is determined as the expansion rate of the target stent on the target cross-section; wherein the first area is the third lumen cross-sectional area, or the cross-sectional area of ​​the target stent after expansion; and the second area is any one of the first lumen cross-sectional area, the second lumen cross-sectional area, the average area of ​​the first lumen cross-sectional area and the second lumen cross-sectional area.

[0050] In a third aspect, an embodiment of the present application further provides an electronic device, comprising: a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the memory communicate through the bus, and when the machine-readable instructions are executed by the processor, the steps in any possible implementation of the first aspect above are performed.

[0051] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps in any possible implementation of the first aspect described above are executed.

[0052] The embodiments of the present application provide a method, device, electronic device and storage medium for predicting the effect of stent implantation, wherein before the stent is actually implanted into the diseased blood vessel, the implantation of each target stent is first performed in a preview manner to predict the implantation effect parameters of the diseased blood vessel and the expansion parameters of the target stent after the target stent is implanted into the diseased blood vessel. After obtaining the implantation effect parameters and expansion parameters corresponding to each target stent, the physician can refer to the implantation effect parameters and expansion parameters of each target stent to select a target stent that is more suitable for the patient. Compared to the prior art method in which physicians select stents based on their own subjective experience, the method of the present application can provide physicians with direct medical reference information (i.e., the implantation effect parameters and expansion parameters of each target stent) when selecting stents, thereby improving the accuracy of physicians in selecting stents.

[0053] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, preferred embodiments are specifically cited below and described in detail with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0055] Figure 1 A flow chart showing a method for predicting stent implantation effects provided in an embodiment of the present application;

[0056] Figure 2 A topological schematic diagram of a stent prediction model provided in an embodiment of the present application is shown;

[0057] Figure 3 A binary image of the distribution position of a target bracket on a target cross section provided by an embodiment of the present application is shown;

[0058] Figure 4 A schematic diagram showing the structure of a device for predicting stent implantation effects provided in an embodiment of the present application is shown;

[0059] Figure 5 A schematic structural diagram of an electronic device provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0060] To make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. The components of the embodiments of the present application generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.

[0061] Considering that in the prior art, when doctors select stents suitable for patients' diseased blood vessels, they can usually only make the selection based on their own subjective experience, which may lead to inappropriate selection of stent parameters for the selected stent, thereby causing the problem of poor stent implantation effect. Based on this, the embodiments of the present application provide a method, device, electronic device and storage medium for predicting the implantation effect of a stent, which can provide doctors with direct medical reference information when selecting stents, which is conducive to improving the accuracy of doctors' selection of stents. It is described below through embodiments.

[0062] Embodiment 1:

[0063] To facilitate understanding of this embodiment, a method for predicting the effect of stent implantation disclosed in an embodiment of the present application is first introduced in detail. Figure 1 A flowchart of a method for predicting the effect of stent implantation provided in an embodiment of the present application is shown. Figure 1 As shown, the following steps are included:

[0064] S101: Obtain an OCT image sequence and an IVUS image sequence for imaging within a diseased blood vessel; the OCT image sequence includes OCT images corresponding to each target section within the diseased blood vessel; the IVUS image sequence includes IVUS images corresponding to each target section.

[0065] Among coronary imaging technologies, intravascular optical coherence tomography (OCT) is an intravascular tomographic imaging method based on optical technology, which can be used to analyze vascular morphology, detect atherosclerotic plaques, and assist in coronary intervention such as stent implantation. Intravascular ultrasound (IVUS) is another intravascular imaging method based on ultrasound technology.

[0066] In this embodiment, the OCT image sequence is obtained by imaging each target cross section in the diseased blood vessel before stent implantation surgery is performed on the diseased blood vessel by intravascular optical coherence tomography imaging method, and the OCT image sequence includes multiple frames of continuous OCT images. The IVUS image sequence is obtained by imaging each target cross section in the diseased blood vessel before stent implantation surgery is performed on the diseased blood vessel by intravascular ultrasound imaging method, and the IVUS image sequence includes multiple frames of continuous IVUS images. The diseased blood vessel is specifically a section of a coronary artery vessel containing a coronary artery lesion.

[0067] S102: Based on the OCT image and IVUS image corresponding to each target section, a lumen region mask, a calcification region mask, and a vascular external elastic membrane region mask corresponding to the target section are generated.

[0068] In this embodiment, the lumen area mask corresponding to the target section is used to represent the lumen area of ​​the diseased blood vessel on the target section, and the lumen area mask can be obtained by filling the lumen area with a code. The calcification area mask corresponding to the target section is used to represent the calcification area of ​​the diseased blood vessel on the target section, and the calcification area mask can be obtained by filling the calcification area with a code. The vascular external elastic membrane area mask corresponding to the target section is used to represent the vascular external elastic membrane area of ​​the diseased blood vessel on the target section, and the vascular external elastic membrane area mask can be obtained by filling the vascular external elastic membrane area with a code. Among them, each target section corresponds to its own lumen area mask, calcification area mask and vascular external elastic membrane area mask. The number of target sections is multiple.

[0069] S103: Input the OCT image sequence, IVUS image sequence, lumen area mask, calcification area mask, vascular extraelastic membrane area mask and stent parameters of the target stent into the trained stent prediction model to predict the implantation effect parameters of the diseased vessel after the target stent is implanted into the diseased vessel and the expansion parameters of the target stent.

[0070] The OCT image sequence, IVUS image sequence, lumen area mask corresponding to each target section, calcification area mask corresponding to each target section, vascular external elastic membrane area mask corresponding to each target section, and stent parameters of the target stent are input into the trained stent prediction model, and the output is used to represent the implantation effect parameters of the diseased vessel when the implantation time reaches the target time after the target stent is pre-implanted into the diseased vessel and the expansion parameters of the target stent. The target time can be any time from 3 months to 10 years, for example, 3 months, 6 months, 12 months, 18 months, 24 months, etc. When the target time is 0, it means that the target stent has just been implanted into the diseased vessel.

[0071] In this embodiment, the stent parameters include: stent cross-sectional dimensions, stent length, stent placement position, expansion balloon size, and expansion pressure. Among them, the stent cross-sectional dimensions can be the diameter dimension or cross-sectional area of ​​the stent cross section, reflecting the size of the target stent after expansion. The stent placement position is specifically the lesion location in the diseased blood vessel. The expansion balloon is placed in the target stent, and the function of the expansion balloon is that when the expansion balloon is inflated so that the expansion balloon bulges, the expansion balloon can drive the net of the target stent to expand. The expansion balloon size indicates the size of the expansion balloon after the expansion balloon bulges. The expansion pressure indicates the inflation pressure when the expansion balloon is inflated.

[0072] The implantation effect parameters include: the third lumen cross-sectional area of ​​the diseased vessel after expansion by the target stent on each target cross-section, the probability of poor adhesion between the target stent and the diseased vessel, the probability of edge dissection of the diseased vessel, and the probability of tissue prolapse of the diseased vessel; the expansion parameter is the expansion rate. Among them, the third lumen cross-sectional area refers to the lumen cross-sectional area of ​​the diseased vessel after the diseased vessel is expanded by the target stent, and each target cross-section corresponds to a third lumen cross-sectional area.

[0073] The probability of poor adhesion refers to the probability of poor adhesion between the expanded target stent and the lesion vessel after the lesion vessel is expanded using the target stent. For example, poor adhesion can be caused by a low expansion pressure or a small cross-sectional size of the stent, which results in poor adhesion between the target stent and the inner wall of the lesion vessel, thereby making it impossible for the target stent to fully expand the lesion vessel.

[0074] The probability of edge dissection indicates the situation that the metal mesh of the target stent cuts the inner wall of the diseased blood vessel after the diseased blood vessel is expanded using the target stent. For example, if the expansion pressure of the target stent is large after expansion, the metal mesh of the target stent is squeezed with the inner wall of the diseased blood vessel, so that the metal mesh is embedded in the inner wall of the diseased blood vessel, and the metal mesh cuts the inner wall of the diseased blood vessel.

[0075] The tissue prolapse probability indicates the tissue prolapse of the inner wall of the diseased blood vessel after the diseased blood vessel is expanded using the target stent. For example, after the target stent is expanded, due to a large expansion pressure or a large cross-sectional size of the stent, the metal mesh of the target stent is squeezed with the inner wall of the diseased blood vessel, so that the tissue on the inner wall of the diseased blood vessel bulges out from the metal mesh, forming tissue prolapse.

[0076] The expansion parameter is the expansion rate of the target stent, wherein the closer the expansion rate is to 100%, the better the expansion of the target stent is; the smaller the expansion rate (far below 100%), the worse the expansion of the target stent is; and the expansion rate exceeding 100% indicates that the expansion of the target stent is over-expanded. For example, when the diseased blood vessel contains a calcified area, the calcified position will affect the expansion of the target stent, resulting in a smaller expansion rate of the target stent at the calcified position.

[0077] In a possible embodiment, the stent parameters of the target stent may be selected based on the physician's subjective experience. For example, the physician selects 10 target stents and, according to the above method, respectively previews the implantation effect parameters of the diseased blood vessels and the expansion parameters of the target stents after each target stent is implanted into the diseased blood vessels, so that the physician can understand what the effect will be after each target stent is used. The physician can select a target stent that is more suitable for the patient from the 10 target stents based on the implantation effect parameters of the diseased blood vessels and the expansion parameters of the target stents corresponding to each of the 10 target stents.

[0078] Coronary imaging information (including OCT image sequences and IVUS image sequences) can not only provide information such as the lumen size of the diseased blood vessels, but also reflect the type of plaques. For example, calcified plaques on the inner wall of diseased blood vessels are relatively hard and may cause serious poor expansion of stents. This is a factor that must be considered in coronary intervention. Pre-treatment and treatment such as atherectomy and shock wave balloons are required before effective stent implantation. Therefore, it is very important to effectively use coronary imaging information to guide stent placement before surgery.

[0079] In this embodiment, by acquiring an OCT image sequence and an IVUS image sequence before implanting the target stent, and rehearsing the implantation effect parameters of the diseased blood vessel and the expansion parameters of the target stent after implanting the target stent into the diseased blood vessel, it is helpful to optimize the stent parameters of the selected target stent and to assist the physician in selecting the best treatment plan. Otherwise, if the stent parameters of the implanted target stent are inappropriate, since the stent cannot be removed from the patient's body, a series of risk consequences that threaten the patient's prognosis and life safety may occur.

[0080] In a possible implementation, the OCT image sequence and the IVUS image sequence are obtained by imaging the same intravascular imaging device during the same invasion of the diseased blood vessel; or, the OCT image sequence and the IVUS image sequence are obtained by imaging at different times;

[0081] When the OCT image sequence and the IVUS image sequence are obtained at different times, the OCT image sequence and the IVUS image sequence are registered in the following manner:

[0082] According to the target cross section corresponding to each OCT image and the target cross section corresponding to each IVUS image, the OCT image and the IVUS image corresponding to the same target cross section are registered.

[0083] In this embodiment, when the intravascular imaging device includes both intravascular optical coherence tomography technology and intravascular ultrasound imaging technology, the same intravascular imaging device can be used to simultaneously collect OCT images and IVUS images of each target cross-section in the diseased blood vessel during the same invasion of the diseased blood vessel to obtain an OCT image sequence and an IVUS image sequence.

[0084] In this embodiment, different intravascular imaging devices may be used, or the same intravascular imaging device may be used at different times to acquire OCT images and IVUS images of target sections in the diseased blood vessel to obtain an OCT image sequence and an IVUS image sequence. At this time, since the OCT image sequence and the IVUS image sequence are acquired at different times, the OCT image and the IVUS image corresponding to the same target section may be registered based on the target sections corresponding to the respective OCT images and the target sections corresponding to the respective IVUS images, so that each frame image of the two intravascular imaging modalities corresponds to the same target section in the diseased blood vessel.

[0085] In a possible real-time manner, when executing step S102 to generate the lumen region mask, calcification region mask, and vascular external elastic membrane region mask corresponding to each target section based on the OCT image and IVUS image corresponding to the target section, the following steps may be specifically performed:

[0086] S1021: For each IVUS image corresponding to the target cross section, a vascular external elastic membrane region mask corresponding to the target cross section is generated according to the vascular external elastic membrane region segmented from the IVUS image.

[0087] For each IVUS image corresponding to the target section, the boundary of the vascular external elastic membrane in the IVUS image is outlined, so as to segment the vascular external elastic membrane region from the IVUS image according to the boundary of the vascular external elastic membrane. Then, the vascular external elastic membrane region is filled with a code to generate a vascular external elastic membrane region mask.

[0088] S1022: For each OCT image corresponding to the target cross section, generate a lumen region mask and a calcification region mask corresponding to the target cross section according to the lumen region and calcification region segmented from the OCT image.

[0089] Similarly, for each target section corresponding to the OCT image, by outlining the lumen boundary and the calcification boundary in the OCT image, the lumen region is segmented from the OCT image according to the lumen boundary, and the calcification region is segmented from the OCT image according to the calcification boundary. Then, different codes are used to fill the lumen region and the calcification region, respectively, to generate a lumen region mask and a calcification region mask.

[0090] S1023: Alternatively, for each target section corresponding to the OCT image and IVUS image, a lumen region mask and a calcification region mask corresponding to the target section are generated according to the lumen region and the calcification region segmented from the OCT image and the IVUS image.

[0091] Alternatively, the lumen region mask and calcification region mask corresponding to the target section may also be generated based on the OCT image and the IVUS image. Specifically, for each OCT image corresponding to the target section, the lumen boundary and the calcification boundary in the OCT image are outlined, thereby segmenting the first lumen region from the OCT image based on the lumen boundary, and segmenting the first calcification region from the OCT image based on the calcification boundary. Then, different codes are used to fill the first lumen region and the first calcification region, respectively, to generate the first lumen region mask and the first calcification region mask.

[0092] And for the IVUS image corresponding to each target section, by outlining the lumen boundary and the calcification boundary in the IVUS image, a second lumen region is segmented from the IVUS image according to the lumen boundary, and a second calcification region is segmented from the IVUS image according to the calcification boundary. Then, different codes are used to fill the second lumen region and the second calcification region, respectively, to generate a second lumen region mask and a second calcification region mask.

[0093] Then, the lumen region mask corresponding to the target section is obtained by fusing the first lumen region mask and the second lumen region mask corresponding to the same target section, and the calcification region mask corresponding to the target section is obtained by fusing the first calcification region mask and the second calcification region mask corresponding to the same target section.

[0094] In a possible implementation, after executing step S102 to generate the lumen region mask, calcification region mask, and vascular external elastic membrane region mask corresponding to each target section based on the OCT image and IVUS image corresponding to the target section, the following steps may be specifically performed:

[0095] S1024: Determine a first lumen cross-sectional area in the proximal reference section and a second lumen cross-sectional area in the distal reference section according to lumen region masks corresponding to the proximal reference section and the distal reference section in the target section, wherein the proximal reference section and the distal reference section are located at two ends of the diseased blood vessel, respectively.

[0096] In this embodiment, the target cross section is each cross section in the diseased blood vessel, there are multiple target cross sections, the target cross section includes a proximal reference cross section and a distal reference cross section, and the other cross sections in the target cross section are located between the proximal reference cross section and the distal reference cross section. Among them, the proximal reference cross section is specifically the target cross section corresponding to the end of the diseased blood vessel close to the aorta, and the distal reference cross section is the target cross section corresponding to the end of the diseased blood vessel far from the aorta.

[0097] The first lumen cross-sectional area in the proximal reference section is determined according to the lumen region mask corresponding to the proximal reference section in the target section, and the second lumen cross-sectional area in the distal reference section is determined according to the lumen region mask corresponding to the distal reference section in the target section.

[0098] Among them, since the proximal reference section and the distal reference section are respectively located at the two ends of the diseased blood vessel, the two ends of the diseased blood vessel are not the lesions (the lesions are located inside the diseased blood vessel), so the first lumen cross-sectional area corresponding to the proximal reference section and the second lumen cross-sectional area corresponding to the distal reference section can be used as the lumen reference cross-sectional area of ​​the normal blood vessel.

[0099] S1025: Input the OCT image sequence, IVUS image sequence, the first lumen cross-sectional area, the second lumen cross-sectional area, the position of the proximal reference section, the position of the distal reference section, and the lumen area mask corresponding to each target section into the trained stent selection model, and output the stent parameters of at least one target stent.

[0100] In this embodiment, the stent parameters of the target stent may also be selected according to the stent selection model, wherein the stent parameters of the target stent output by the stent selection model are a combination of stent parameters of the target stent suitable for use in the diseased blood vessel. For example, the lumen diameter of the diseased blood vessel is 3 mm, so the stent cross-sectional size in the stent parameters of the output target stent may be 2.5 mm, 3 mm, 3.5 mm, but cannot be 10 mm.

[0101] In a possible implementation, when executing step S103, the OCT image sequence, the IVUS image sequence, the lumen area mask, the calcification area mask, the vascular external elastic membrane area mask, and the stent parameters of the target stent are input into the trained stent prediction model, and the implantation effect parameters of the diseased blood vessel after the target stent is implanted into the diseased blood vessel are predicted, which can be specifically performed according to the following steps:

[0102] S1031: The OCT image sequence, the IVUS image sequence, the lumen region mask, the calcification region mask, and the vascular external elastic membrane region mask are respectively generated through the corresponding first encoders and first decoders to generate the corresponding first feature matrices.

[0103] Figure 2 A topological schematic diagram of a stent prediction model provided in an embodiment of the present application is shown, such as Figure 2 As shown, the OCT image sequence passes through the first encoder A1 and the first decoder B1 in sequence to generate the first characteristic matrix C1 corresponding to the OCT image sequence. The IVUS image sequence passes through the first encoder A2 and the first decoder B2 in sequence to generate the first characteristic matrix C2 corresponding to the IVUS image sequence. The lumen area mask corresponding to each target section passes through the first encoder A3 and the first decoder B3 in sequence to generate the first characteristic matrix C3 corresponding to all lumen area masks. The calcification area mask corresponding to each target section passes through the first encoder A4 and the first decoder B4 in sequence to generate the first characteristic matrix C4 corresponding to all calcification area masks. The vascular external elastic membrane area mask corresponding to each target section passes through the first encoder A5 and the first decoder B5 in sequence to generate the first characteristic matrix C5 corresponding to all vascular external elastic membrane area masks.

[0104] S1032: Passing a second feature matrix obtained by fusing the first feature matrix through a second encoder to generate a third feature matrix.

[0105] The first feature matrix C1, the first feature matrix C2, the first feature matrix C3, the first feature matrix C4, and the first feature matrix C5 are fused to obtain a second feature matrix. The second feature matrix is ​​passed through a second encoder to generate a third feature matrix.

[0106] S1033: Input the third feature matrix and the stent parameters into the fully connected layer, output the third lumen cross-sectional area of ​​the diseased blood vessel on each target cross-section after expansion by the target stent; and generate a fourth feature matrix from the third feature matrix through the second decoder.

[0107] The third characteristic matrix and the stent parameters are input into one or more fully connected layers, and the third lumen cross-sectional area of ​​the diseased blood vessel on each target cross-section after expansion by the target stent is output.

[0108] S1034: The fifth feature matrix obtained by fusing the fourth feature matrix, the pressure feature matrix and the stent cross-sectional size matrix is ​​input into the convolution layer, and the probability of poor adhesion between the target stent and the diseased vessel, the probability of edge dissection of the diseased vessel and the probability of tissue prolapse of the diseased vessel on each target section are output.

[0109] The pressure feature matrix is ​​used to represent a heat map representing the pressure magnitude based on the selected target stent cross-sectional size, expansion balloon size, and expansion pressure, reflecting the pressure distribution of the corresponding diseased blood vessel expansion position after the target stent is expanded. The stent cross-sectional size matrix is ​​used to represent a binary map representing the stent distribution position of the target stent on each target cross section after the target stent is expanded based on the selected target stent cross-sectional size, expansion balloon size, and expansion pressure. Figure 3 The binary image of the distribution position of the target bracket provided in the embodiment of the present application on a certain target cross section is shown, such as Figure 3 As shown, the black circle represents the expanded distribution position of the target stent on one of the target cross sections after the target stent is expanded. In this embodiment, the role of the encoder (including the first encoder and the second encoder) is to encode and abstract the input feature matrix layer by layer through several layers of convolution, nonlinear activation, and downsampling operations. The role of the decoder (including the first decoder and the second decoder) is to gradually restore the feature matrix output by the encoder through several layers of convolution, nonlinear activation, upsampling or deconvolution operations. The depth of the encoder and decoder can be the same or different. The encoder and decoder can be a convolutional neural network, or a Transformer model, or a combination of a convolutional neural network and a Transformer model.

[0110] In a possible implementation manner, when executing step S103 to predict the expansion parameters of the target stent after the target stent is implanted into the diseased blood vessel, the following steps may be specifically performed:

[0111] For each target cross-section, the ratio of the first area corresponding to the target cross-section to the second area is determined as the expansion rate of the target stent on the target cross-section; wherein the first area is the third lumen cross-sectional area, or the stent cross-sectional area after the target stent is expanded; the second area is any one of the first lumen cross-sectional area, the second lumen cross-sectional area, and the average area of ​​the first lumen cross-sectional area and the second lumen cross-sectional area.

[0112] In this embodiment, the expansion rate of the target stent is calculated by the following formula:

[0113]

[0114] Where i represents the i-th target section; P i represents the expansion rate of the target bracket corresponding to the i-th target section; S i S represents the third lumen cross-sectional area corresponding to the i-th target cross-sectional area, or the cross-sectional area of ​​the target stent after expansion corresponding to the i-th target cross-sectional area; refIt represents the first lumen cross-sectional area, or the second lumen cross-sectional area, or the average area of ​​the first lumen cross-sectional area and the second lumen cross-sectional area.

[0115] Embodiment 2:

[0116] Based on the same technical concept, the present application also provides a device for predicting the effect of stent implantation. Figure 4 A schematic diagram of the structure of a prediction of the effect of stent implantation provided in an embodiment of the present application is shown, such as Figure 4 As shown, the device comprises:

[0117] An acquisition module 401 is used to acquire an OCT image sequence and an IVUS image sequence for imaging within a diseased blood vessel; the OCT image sequence includes OCT images corresponding to each target section within the diseased blood vessel; the IVUS image sequence includes IVUS images corresponding to each target section;

[0118] A generating module 402, for generating a lumen region mask, a calcification region mask, and a vascular external elastic membrane region mask corresponding to each target section based on the OCT image and the IVUS image corresponding to the target section;

[0119] The prediction module 403 is used to input the OCT image sequence, the IVUS image sequence, the lumen area mask, the calcification area mask, the vascular extraelastic membrane area mask and the stent parameters of the target stent into the trained stent prediction model, and predict the implantation effect parameters of the diseased blood vessel after the target stent is implanted into the diseased blood vessel and the expansion parameters of the target stent.

[0120] Optionally, the OCT image sequence and the IVUS image sequence are obtained by imaging the same intravascular imaging device during the same invasion of the diseased blood vessel; or, the OCT image sequence and the IVUS image sequence are obtained by imaging at different times;

[0121] When the OCT image sequence and the IVUS image sequence are obtained by imaging at different times, the OCT image sequence and the IVUS image sequence are registered in the following manner:

[0122] According to the target cross section corresponding to each of the OCT images and the target cross section corresponding to each of the IVUS images, the OCT image and the IVUS image corresponding to the same target cross section are registered.

[0123] Optionally, when the generating module 402 is used to generate the lumen region mask, the calcification region mask, and the vascular external elastic membrane region mask corresponding to each target section based on the OCT image and the IVUS image corresponding to the target section, it is specifically used to:

[0124] For each IVUS image corresponding to the target section, generating a mask of the vascular external elastic membrane region corresponding to the target section according to the vascular external elastic membrane region segmented from the IVUS image;

[0125] For each of the OCT images corresponding to the target cross section, generating the lumen region mask and the calcification region mask corresponding to the target cross section according to the lumen region and the calcification region segmented from the OCT image;

[0126] Alternatively, for the OCT image and the IVUS image corresponding to each target section, the lumen region mask and the calcification region mask corresponding to the target section are generated according to the lumen region and the calcification region segmented from the OCT image and the IVUS image.

[0127] Optionally, also include:

[0128] A determination module, configured to determine a first lumen cross-sectional area in the proximal reference section and a second lumen cross-sectional area in the distal reference section according to the lumen region masks respectively corresponding to the proximal reference section and the distal reference section in the target section, after the generation module 402 is configured to generate a lumen region mask, a calcification region mask, and a vascular external elastic membrane region mask corresponding to the target section based on the OCT image and the IVUS image corresponding to each target section, wherein the proximal reference section and the distal reference section are respectively located at two ends of the diseased blood vessel;

[0129] An input module is used to input the OCT image sequence, the IVUS image sequence, the first lumen cross-sectional area, the second lumen cross-sectional area, the position of the proximal reference section, the position of the distal reference section, and the lumen area mask corresponding to each target section into a trained stent selection model, and output the stent parameters of at least one of the target stents.

[0130] Optionally, the stent parameters include: stent cross-sectional size, stent length, stent placement position, expansion balloon size, and expansion pressure; the implantation effect parameters include: the third lumen cross-sectional area of ​​the diseased vessel on each target cross-section after expansion by the target stent, the probability of poor adhesion between the target stent and the diseased vessel, the probability of edge dissection of the diseased vessel, and the probability of tissue prolapse of the diseased vessel; the expansion parameter is the expansion rate.

[0131] Optionally, the prediction module 403 is used to input the OCT image sequence, the IVUS image sequence, the lumen area mask, the calcification area mask, the vascular external elastic membrane area mask, and the stent parameters of the target stent into the trained stent prediction model, and predict the implantation effect parameters of the diseased blood vessel after the target stent is implanted into the diseased blood vessel, specifically for:

[0132] The OCT image sequence, the IVUS image sequence, the lumen region mask, the calcification region mask, and the vascular external elastic membrane region mask are respectively generated through the first encoder and the first decoder corresponding to each other to form the first feature matrix corresponding to each other;

[0133] Passing a second feature matrix obtained by fusing the first feature matrix through a second encoder to generate a third feature matrix;

[0134] Input the third characteristic matrix and the stent parameters into the fully connected layer, output the third lumen cross-sectional area of ​​the diseased blood vessel on each target cross-section after expansion by the target stent; and generate a fourth characteristic matrix by using the third characteristic matrix through a second decoder;

[0135] The fifth characteristic matrix obtained by fusing the fourth characteristic matrix, the pressure characteristic matrix and the stent cross-sectional size matrix is ​​input into the convolution layer, and the probability of poor adhesion between the target stent and the diseased blood vessel, the probability of edge dissection of the diseased blood vessel and the probability of tissue prolapse of the diseased blood vessel on each target section are output.

[0136] Optionally, when the prediction module 403 is used to predict the expansion parameters of the target stent after the target stent is implanted into the diseased blood vessel, it is specifically used to:

[0137] For each target cross-section, the ratio of the first area to the second area corresponding to the target cross-section is determined as the expansion rate of the target stent on the target cross-section; wherein the first area is the third lumen cross-sectional area, or the cross-sectional area of ​​the target stent after expansion; and the second area is any one of the first lumen cross-sectional area, the second lumen cross-sectional area, the average area of ​​the first lumen cross-sectional area and the second lumen cross-sectional area.

[0138] Embodiment three:

[0139] Based on the same technical concept, the embodiment of the present application also provides an electronic device, Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application is shown. Figure 5As shown, the electronic device 500 includes: a processor 501, a memory 502 and a bus 503. The memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor 501 communicates with the memory 502 through the bus 503, and the processor 501 executes the machine-readable instructions to perform the method steps described in Example 1.

[0140] Embodiment 4:

[0141] Based on the same technical concept, embodiment 4 of the present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the method steps described in embodiment 1 are executed.

[0142] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described devices, electronic devices, and computer-readable storage media can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0143] In the several embodiments provided in the present application, it should be understood that the disclosed methods, devices, electronic devices and computer-readable storage media can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation. For example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of devices or modules can be electrical, mechanical or other forms.

[0144] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, that is, they may be located in one place or distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0145] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0146] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that is executable by a processor. Based on this understanding, the technical solution of the present application can essentially be embodied in the form of a software product, or in other words, the part that contributes to the prior art or the part of the technical solution. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0147] Finally, it should be noted that the above-described embodiments are only specific implementation methods of the present application, which are used to illustrate the technical solutions of the present application, rather than to limit them. The protection scope of the present application is not limited thereto. Although the present application is described in detail with reference to the above-described embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-described embodiments within the technical scope disclosed in the present application, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A method for predicting the effect of stent implantation, characterized in that: include: Acquire an OCT image sequence and an IVUS image sequence for imaging within a diseased blood vessel; the OCT image sequence includes OCT images corresponding to each target section within the diseased blood vessel; the IVUS image sequence includes IVUS images corresponding to each target section; Based on the OCT image and the IVUS image corresponding to each target section, a lumen region mask, a calcification region mask, and a vascular external elastic membrane region mask corresponding to the target section are generated; Inputting the OCT image sequence, the IVUS image sequence, the lumen region mask, the calcification region mask, the vascular external elastic membrane region mask, and the stent parameters of the target stent into the trained stent prediction model, and predicting the implantation effect parameters of the diseased blood vessel after the target stent is implanted into the diseased blood vessel and the expansion parameters of the target stent; The implantation effect parameters include: the third lumen cross-sectional area of ​​the diseased blood vessel after expansion by the target stent at each target cross-section, the probability of poor adhesion between the target stent and the diseased blood vessel, the probability of edge dissection of the diseased blood vessel, and the probability of tissue prolapse of the diseased blood vessel; The OCT image sequence, the IVUS image sequence, the lumen region mask, the calcification region mask, the vascular external elastic membrane region mask, and the stent parameters of the target stent are input into a trained stent prediction model to predict the implantation effect parameters of the diseased blood vessel after the target stent is implanted into the diseased blood vessel, including: The OCT image sequence, the IVUS image sequence, the lumen region mask, the calcification region mask, and the vascular external elastic membrane region mask are respectively generated through the first encoder and the first decoder corresponding to each other to form the first feature matrix corresponding to each other; Passing a second feature matrix obtained by fusing the first feature matrix through a second encoder to generate a third feature matrix; Input the third characteristic matrix and the stent parameters into the fully connected layer, output the third lumen cross-sectional area of ​​the diseased blood vessel on each target cross-section after expansion by the target stent; and generate a fourth characteristic matrix by using the third characteristic matrix through a second decoder; The fifth characteristic matrix obtained by fusing the fourth characteristic matrix, the pressure characteristic matrix and the stent cross-sectional size matrix is ​​input into the convolution layer, and the probability of poor adhesion between the target stent and the diseased blood vessel, the probability of edge dissection of the diseased blood vessel and the probability of tissue prolapse of the diseased blood vessel on each target section are output.

2. The method according to claim 1, characterized in that: The OCT image sequence and the IVUS image sequence are obtained by imaging the same intravascular imaging device during the same invasion of the diseased blood vessel; or, the OCT image sequence and the IVUS image sequence are obtained by imaging at different times; When the OCT image sequence and the IVUS image sequence are obtained by imaging at different times, the OCT image sequence and the IVUS image sequence are registered in the following manner: According to the target cross-section corresponding to each of the OCT images and the target cross-section corresponding to each of the IVUS images, the OCT image and the IVUS image corresponding to the same target cross-section are registered.

3. The method according to claim 1, characterized in that: The step of generating a lumen region mask, a calcification region mask, and a vascular external elastic membrane region mask corresponding to each target section based on the OCT image and the IVUS image corresponding to the target section includes: For each IVUS image corresponding to the target section, generating a mask of the vascular external elastic membrane region corresponding to the target section according to the vascular external elastic membrane region segmented from the IVUS image; For each of the OCT images corresponding to the target cross section, generating the lumen region mask and the calcification region mask corresponding to the target cross section according to the lumen region and the calcification region segmented from the OCT image; Alternatively, for the OCT image and the IVUS image corresponding to each target section, the lumen region mask and the calcification region mask corresponding to the target section are generated according to the lumen region and the calcification region segmented from the OCT image and the IVUS image.

4. The method according to claim 1, characterized in that: After generating the lumen region mask, the calcification region mask, and the vascular external elastic membrane region mask corresponding to each target section based on the OCT image and the IVUS image corresponding to the target section, the method further includes: Determine a first lumen cross-sectional area in the proximal reference section and a second lumen cross-sectional area in the distal reference section according to the lumen region masks respectively corresponding to the proximal reference section and the distal reference section in the target section; wherein the proximal reference section and the distal reference section are respectively located at two ends of the diseased blood vessel; The OCT image sequence, the IVUS image sequence, the first lumen cross-sectional area, the second lumen cross-sectional area, the position of the proximal reference section, the position of the distal reference section, and the lumen area mask corresponding to each target section are input into the trained stent selection model, and the stent parameters of at least one of the target stents are output.

5. The method according to claim 4, characterized in that: The stent parameters include: stent cross-sectional size, stent length, stent placement position, expansion balloon size, and expansion pressure; the expansion parameter is the expansion rate.

6. The method according to claim 5, characterized in that: The predicting of expansion parameters of the target stent after the target stent is implanted into the diseased blood vessel includes: For each target cross-section, the ratio of the first area to the second area corresponding to the target cross-section is determined as the expansion rate of the target stent on the target cross-section; wherein the first area is the third lumen cross-sectional area, or the cross-sectional area of ​​the target stent after expansion; and the second area is any one of the first lumen cross-sectional area, the second lumen cross-sectional area, the average area of ​​the first lumen cross-sectional area and the second lumen cross-sectional area.

7. A device for predicting the effect of stent implantation, characterized in that: include: An acquisition module, used to acquire an OCT image sequence and an IVUS image sequence for imaging within a diseased blood vessel; the OCT image sequence includes OCT images corresponding to each target section within the diseased blood vessel; the IVUS image sequence includes IVUS images corresponding to each target section; A generating module, for generating a lumen region mask, a calcification region mask, and a vascular external elastic membrane region mask corresponding to each target section based on the OCT image and the IVUS image corresponding to the target section; A prediction module, used for inputting the OCT image sequence, the IVUS image sequence, the lumen region mask, the calcification region mask, the vascular external elastic membrane region mask and the stent parameters of the target stent into the trained stent prediction model, and predicting the implantation effect parameters of the diseased blood vessel and the expansion parameters of the target stent after the target stent is implanted into the diseased blood vessel; The implantation effect parameters include: the third lumen cross-sectional area of ​​the diseased blood vessel after expansion by the target stent at each target cross-section, the probability of poor adhesion between the target stent and the diseased blood vessel, the probability of edge dissection of the diseased blood vessel, and the probability of tissue prolapse of the diseased blood vessel; The prediction module is used to input the OCT image sequence, the IVUS image sequence, the lumen area mask, the calcification area mask, the vascular external elastic membrane area mask and the stent parameters of the target stent into the trained stent prediction model, and predict the implantation effect parameters of the diseased blood vessel after the target stent is implanted into the diseased blood vessel, specifically for: The OCT image sequence, the IVUS image sequence, the lumen region mask, the calcification region mask, and the vascular external elastic membrane region mask are respectively generated through the first encoder and the first decoder corresponding to each other to form the first feature matrix corresponding to each other; Passing a second feature matrix obtained by fusing the first feature matrix through a second encoder to generate a third feature matrix; Input the third characteristic matrix and the stent parameters into the fully connected layer, output the third lumen cross-sectional area of ​​the diseased blood vessel on each target cross-section after expansion by the target stent; and generate a fourth characteristic matrix by using the third characteristic matrix through a second decoder; The fifth characteristic matrix obtained by fusing the fourth characteristic matrix, the pressure characteristic matrix and the stent cross-sectional size matrix is ​​input into the convolution layer, and the probability of poor adhesion between the target stent and the diseased blood vessel, the probability of edge dissection of the diseased blood vessel and the probability of tissue prolapse of the diseased blood vessel on each target section are output.

8. An electronic device, characterized in that: include: A processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the memory communicate via the bus, and when the machine-readable instructions are executed by the processor, the steps of the method as described in any one of claims 1 to 6 are performed.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are executed.

Images