CN116958692A

CN116958692A - Angiography image key frame selection method, angiography image key frame selection device, angiography image key frame selection equipment and angiography image key frame selection medium

Info

Publication number: CN116958692A
Application number: CN202310940710.1A
Authority: CN
Inventors: 高丰伟; 王季勇; 陈树湛
Original assignee: Shanghai Bodong Medical Technology Co ltd
Current assignee: Shanghai Bodong Medical Technology Co ltd
Priority date: 2023-07-27
Filing date: 2023-07-27
Publication date: 2023-10-27

Abstract

The embodiment of the invention discloses a method, a device, equipment and a medium for selecting an angiography image key frame. The method comprises the following steps: acquiring an angiography sequence, and identifying vascular stenosis of the angiography images aiming at each angiography image in the angiography sequence; based on the vascular stenosis recognition results respectively corresponding to each angiography image, and according to a preset key frame selection strategy, a single key frame is selected from the angiography sequence. According to the technical scheme provided by the embodiment of the invention, the key frames can be automatically selected from the angiography sequence, so that the workload of doctors is reduced.

Description

Angiography image key frame selection method, angiography image key frame selection device, angiography image key frame selection equipment and angiography image key frame selection medium

Technical Field

The embodiment of the invention relates to the field of image processing, in particular to a method, a device, equipment and a medium for selecting an angiography image key frame.

Background

In recent years, the incidence of cardiovascular diseases has been on the rise, and people pay more attention. Coronary heart disease is one of cardiovascular diseases, and mainly refers to a series of clinical syndromes caused by coronary artery stenosis, such as palpitation, shortness of breath, angina pectoris, myocardial infarction and the like.

Clinically, the "gold standard" for accurately detecting coronary artery stenosis is the fractional flow reserve (Fraction Flow Reservation, FFR) of the coronary artery, but this detection scheme has the problems of high cost of manpower and materials and limited applicable population. In order to solve the above-mentioned problems, an implementation scheme is proposed that a key frame is selected from a digital subtraction angiography (Digital subtraction angiography, DSA) sequence, and then coronary artery stenosis (i.e. vascular stenosis) detection is performed based on the key frame. For simplicity of description, digital subtraction angiography may be referred to simply as angiography.

It should be noted that, currently, key frames are selected manually by a doctor. Obviously, this key frame selection scheme increases the workload of doctors and needs to be improved.

Disclosure of Invention

The embodiment of the invention provides a method, a device, equipment and a medium for selecting key frames of angiography images, which are used for realizing automatic selection of the key frames from angiography sequences, so that the workload of doctors is reduced.

According to an aspect of the present invention, there is provided an angiographic image key frame selection method, which may include:

acquiring an angiography sequence, and identifying vascular stenosis of the angiography images aiming at each angiography image in the angiography sequence;

Based on the vascular stenosis recognition results corresponding to each angiography image, a single key frame is selected from the angiography sequence according to a preset key frame selection strategy.

Optionally, the number of key frame selection policies is at least one;

based on the recognition result of the vascular stenosis corresponding to each angiography image, and according to a preset key frame selection strategy, selecting a single key frame from the angiography sequence, wherein the method comprises the following steps:

selecting a candidate sequence from the angiography sequence according to the key frame selection strategy based on the vascular stenosis recognition result corresponding to each angiography image;

and processing the obtained at least one candidate sequence, and selecting a single key frame from angiography images respectively contained in the at least one candidate sequence according to a processing result.

On the basis, according to the processing result, selecting a single key frame from angiographic images respectively contained in at least one candidate sequence, wherein the single key frame comprises the following steps:

selecting at least one candidate frame from angiographic images respectively contained in at least one candidate sequence according to the obtained processing result;

Aiming at each candidate frame in at least one candidate frame, obtaining the vessel diameter of the candidate frame at the vessel stenosis according to the vessel stenosis identification result corresponding to the candidate frame;

and taking the candidate frame with the smallest vessel diameter in the at least one candidate frame as a key frame.

On the basis, optionally, obtaining the vessel diameter of the candidate frame at the vessel stenosis according to the vessel stenosis identification result corresponding to the candidate frame, including:

according to the recognition result of the vascular stenosis corresponding to the candidate frame, determining the central line of the candidate frame at the vascular stenosis, and obtaining two intersection points of the perpendicular line of the central line on the vascular boundary of the vascular stenosis;

the distance between the two intersection points is used as the diameter of the blood vessel of the candidate frame at the narrow part of the blood vessel.

Alternatively, the processing is performed on the obtained at least one candidate sequence, and according to the processing result, a single key frame is selected from angiographic images respectively contained in the at least one candidate sequence, including:

the method comprises the steps of obtaining weights corresponding to at least one obtained candidate sequence respectively, and carrying out assignment processing on the at least one candidate sequence according to the weights corresponding to the at least one candidate sequence respectively;

and selecting a single key frame from angiography images respectively contained in at least one candidate sequence according to the obtained assignment processing result.

On the basis, optionally, assignment processing is performed on at least one candidate sequence according to weights respectively corresponding to the at least one candidate sequence, including:

and for each candidate sequence in the at least one candidate sequence, respectively carrying out assignment processing on each angiographic image in the candidate sequence according to the weight corresponding to the candidate sequence.

On the basis, optionally, under the condition that each angiography image in the candidate sequence is respectively subjected to assignment processing, selecting a single key frame from angiography images respectively contained in at least one candidate sequence according to the obtained assignment processing result, wherein the method comprises the following steps of:

combining two or more assignment processing results corresponding to the repeated images in all angiographic images contained in at least one candidate sequence, wherein the repeated images exist in the two or more candidate sequences, and taking the combination result as the assignment processing result of the repeated images;

and selecting a single key frame from all the angiography images according to the assignment processing results respectively corresponding to all the angiography images.

Optionally, the key frame selection policy is determined based on key frame requirements, wherein the key frame requirements characterize requirements that the key frame for vascular stenosis detection needs to meet.

On this basis, optionally, the class number of the key frame requirements includes at least one class, and each class of key frame requirements corresponds to different key frame selection strategies.

On this basis, optionally, the at least one type of key frame requirement includes at least one of a first type of requirement, a second type of requirement, and a third type of requirement;

wherein the first class of requirements includes at least one of intravascular contrast medium filling, vessel in a relaxed state, and vessel no foreshortening at a vessel stenosis;

the second category of requirements includes the absence of overlapping vessels at a vascular stenosis;

a third class of requirements includes at least one of the absence of artifacts at the vascular stenosis and the clear boundaries of the blood vessel at the vascular stenosis.

Optionally, for each angiographic image in the angiographic sequence, further comprising:

and performing vessel segmentation on the angiography image to obtain a vessel segmentation image.

On the basis, optionally, based on the recognition result of the angiostenosis corresponding to each angiography image, and according to a preset key frame selection strategy, selecting a single key frame from the angiography sequence, including:

based on the blood vessel stenosis recognition result and the blood vessel segmentation image which are respectively corresponding to each angiography image, a single key frame is selected from the angiography sequence according to a preset key frame selection strategy.

According to another aspect of the present invention, there is provided an angiographic image key frame selection device, which may include:

the angiostenosis recognition module is used for acquiring an angiography sequence and recognizing angiostenosis of the angiography images aiming at each angiography image in the angiography sequence;

the key frame selection module is used for selecting a single key frame from the angiography sequence based on the vascular stenosis identification result corresponding to each angiography image respectively and according to a preset key frame selection strategy.

According to another aspect of the present invention, there is provided an electronic device, which may include:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein,,

the memory stores a computer program executable by the at least one processor, the computer program being executable by the at least one processor to cause the at least one processor to implement the angiographic image keyframe selection method provided by any embodiment of the invention when executed.

According to another aspect of the present invention, there is provided a computer readable storage medium having stored thereon computer instructions for causing a processor to perform the angiographic image key-frame selection method provided by any embodiment of the present invention when executed.

According to the technical scheme, the angiography images are subjected to vascular stenosis identification by acquiring the angiography sequence and aiming at each angiography image in the angiography sequence; then, based on the vascular stenosis recognition results corresponding to each angiography image, a single key frame is selected from the angiography sequence according to a preset key frame selection strategy. According to the technical scheme, the key frames can be automatically selected from the angiography sequence, so that the workload of doctors is reduced.

It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the invention, nor is it intended to be used to limit the scope of the invention. Other features of the present invention will become apparent from the description that follows.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a flow chart of a method for key frame selection of angiographic images according to an embodiment of the invention;

FIG. 2 is a flowchart of an example of identifying a stenosis in an angiographic image key frame selection method according to an embodiment of the present invention;

FIG. 3 is a schematic diagram of a recognition result of a vascular stenosis in an angiography image key frame selection method according to an embodiment of the present invention;

FIG. 4 is a flowchart of an example of vessel segmentation in an angiographic image key frame selection method according to an embodiment of the invention;

FIG. 5 is a flow chart of another angiographic image keyframe selection method provided according to an embodiment of the invention;

FIG. 6 is a flow chart of yet another angiographic image key frame selection method provided in accordance with an embodiment of the invention;

FIG. 7 is a schematic diagram of an example of vessel diameter determination in yet another angiographic image key frame selection method provided in accordance with an embodiment of the invention;

FIG. 8 is a flow chart of yet another angiographic image key frame selection method provided in accordance with an embodiment of the invention;

FIG. 9 is a flowchart of an alternative example of a method for key frame selection of an angiographic image, provided in accordance with an embodiment of the invention;

FIG. 10 is a block diagram of an angiographic image key frame selection device according to an embodiment of the invention;

fig. 11 is a schematic structural diagram of an electronic device for implementing the angiographic image key frame selection method in an embodiment of the invention.

Detailed Description

In order that those skilled in the art will better understand the present invention, a technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.

It should be noted that the terms "first," "second," and the like in the description and the claims of the present invention and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the invention described herein may be implemented in sequences other than those illustrated or otherwise described herein. The cases of "target", "original", etc. are similar and will not be described in detail herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

Fig. 1 is a flowchart of a method for selecting a keyframe of an angiographic image according to an embodiment of the invention. The present embodiment may be applied to the case of selecting key frames from an angiographic sequence that can be used to detect vascular stenosis. The method can be implemented by the angiographic image key frame selection device provided by the embodiment of the invention, the device can be implemented by software and/or hardware, and the device can be integrated on electronic equipment, and the electronic equipment can be various user terminals or servers.

Referring to fig. 1, the method of the embodiment of the present invention specifically includes the following steps:

s110, acquiring an angiography sequence, and identifying vascular stenosis of each angiography image in the angiography sequence.

Wherein an angiographic sequence is acquired, which may comprise one or more angiographic images, and which may be combined with an application scenario according to an embodiment of the present invention, in particular a plurality of angiographic images obtained in succession. It should be noted that, the angiographic image may be an angiographic image obtained originally; may be an angiographic image (i.e. a vessel segmentation image or a vessel mask image) in which vessel segmentation has been completed; the angiographic image can be obtained originally and can be directly used as a vessel segmentation image; etc., and are not specifically treated herein.

And carrying out vascular stenosis recognition on the angiography images aiming at each angiography image in the angiography sequence to obtain a vascular stenosis recognition result. In practical applications, alternatively, the recognition of the vascular stenosis (i.e. the detection of the vascular stenosis) may be implemented by various methods, for example, a conventional target recognition method may be exemplified, for example, a method for recognizing the vascular stenosis by performing pattern recognition using target features, wherein the target features may be, for example, scale-invariant feature transform (Scale-invariant feature transform, SIFT) or direction gradient histogram (histogram of oriented gradients, HOG), etc., and the pattern recognition may be, for example, a support vector machine (Support Vector Machine, SVM) or random forest, etc.; the target recognition method based on deep learning can be, for example, RCNN series, YOLO series or the like; of course, the remaining methods for identifying vascular stenosis may be implemented, and are not particularly limited herein.

Illustratively, taking the method of implementing the recognition of the angiostenosis as the angiostenosis recognition module, in practical application, optionally, as shown in fig. 2, consecutive frames (i.e. consecutive multiple angiographic images) in the angiographic sequence are input into the angiostenosis recognition module, so that the recognition of the angiostenosis can be performed on the angiographic images respectively by using the angiostenosis recognition module; further, according to the output result of the vascular stenosis recognition module, the vascular stenosis recognition results corresponding to the angiography images can be obtained. Still further exemplary, the vessel stenosis recognition result may be marked by a position frame, for example, a rectangular frame shown in fig. 3 is a position frame, and a region framed by the position frame is a vessel stenosis.

S120, based on the vascular stenosis recognition results corresponding to each angiography image, selecting a single key frame from the angiography sequence according to a preset key frame selection strategy.

The key frame selection strategy is obtained through preset, and can be understood as a strategy for selecting key frames from an angiography sequence, and particularly can be understood as a strategy for selecting key frames capable of realizing vascular stenosis detection from the angiography sequence. And acquiring a key frame selection strategy.

On the basis, further, based on the vascular stenosis recognition results respectively corresponding to each angiography image and combining a key frame selection strategy, a single angiography image can be selected from the angiography sequence, particularly, a single angiography image which can be used for realizing vascular stenosis detection is selected, and then the selected angiography image is used as a key frame for application.

An alternative solution is to determine a key frame selection strategy according to key frame requirements, wherein the key frame requirements characterize requirements to be met by a key frame for vascular stenosis detection.

The key frame requirement is understood as a requirement to be satisfied by a key frame for realizing the detection of the vascular stenosis. The key frame selection strategy can be predetermined according to the key frame requirement, and the advantage of the arrangement is that the accurate detection of the vascular stenosis can be realized based on the key frames selected by the key frame selection strategy.

In other words, in practical application, there may be one, two or more types of key frame requirements, and the focus of each type of key frame requirement is different, so that key frame selection policies may be set for each type of key frame requirement respectively. The method has the advantages that key frames which are accurately matched with corresponding key frame requirements can be respectively selected based on each key frame selection strategy, key frames meeting the key frame requirements can be further obtained based on the key frames, and the finally obtained key frames can realize accurate detection of vascular stenosis.

wherein the first class of requirements includes at least one of intravascular contrast medium filling, vessel in a relaxed state, and vessel no foreshortening at a vessel stenosis; the second category of requirements includes the absence of overlapping vessels at a vascular stenosis; a third class of requirements includes at least one of the absence of artifacts at the vascular stenosis and the clear boundaries of the blood vessel at the vascular stenosis.

In practical application, the key frame selection strategy determined according to the first type of requirement can select a key frame through a first area of a first blood vessel contained in the angiography image and/or a second area of a second blood vessel contained in the angiography image at a blood vessel stenosis position.

The second type of requirement focuses on the position of the blood vessel, and in practical application, optionally, a key frame selection strategy determined according to the second type of requirement can select a key frame according to the bifurcation condition of the center line in the second blood vessel.

The third type of requirement focuses on the definition of the blood vessel, and in practical application, optionally, a key frame is selected according to a key frame selection strategy determined by the third type of requirement, wherein the key frame can be selected according to the change condition of the pixel values of all pixel points on the vertical line of the central line in the second blood vessel.

In another optional aspect, after each angiographic image in the angiographic sequence, the angiographic image keyframe selection method further includes:

In the case where the angiographic image is an angiographic image which is originally acquired and cannot be directly applied as a vessel segmentation image, the vessel segmentation image (i.e., a vessel mask image) may be obtained by vessel segmentation of the angiographic image, so that a subsequent process may be performed on the vessel segmentation image. It can be appreciated that the vessel segmentation process described above may improve the convenience and accuracy of subsequent processing, since the vessel segmentation image contains only vessels and does not contain the rest. In practical applications, the above-mentioned blood vessel segmentation process may be implemented in various ways, and exemplary blood vessel segmentation may be performed by using a conventional segmentation algorithm, for example, a Max-flow min-cut (Max-flow min-cut), a level set, or a conditional random field; the segmentation algorithm based on deep learning can be utilized to segment blood vessels, for example, the segmentation algorithm can be a unet, a vnet or a nnunet; the existing segmentation software can also be used for carrying out blood vessel segmentation; etc., and are not particularly limited herein.

For example, taking the method of implementing vessel segmentation as a vessel segmentation module, in practical application, optionally, as shown in fig. 4, consecutive frames (i.e. a plurality of angiographic images obtained continuously) in an angiographic sequence are input into the vessel segmentation module, so that vessel segmentation can be performed on the angiographic images respectively by using the vessel segmentation module; further, according to the output result of the blood vessel segmentation module, blood vessel segmentation images corresponding to the angiography images are obtained.

That is, the vascular stenosis recognition result and the vascular segmentation image may be simultaneously applied to select the key frame. Illustratively, according to the vessel stenosis recognition result of the angiographic image, a region of interest (region of interest, ROI) may be selected from the vessel segmentation image corresponding to the angiographic image, i.e., a region corresponding to the vessel stenosis recognition result in the vessel segmentation image is selected as the ROI. The ROI may then be utilized and key frames selected according to a key frame selection policy.

According to the technical scheme, the blood vessel stenosis recognition result is matched with the blood vessel segmentation image to select the key frame, and the application of the blood vessel segmentation image only containing blood vessels improves the accuracy of key frame selection.

Fig. 5 is a flowchart of another angiographic image keyframe selection method provided in an embodiment of the invention. The present embodiment is optimized based on the above technical solutions. In this embodiment, optionally, the number of key frame selection policies is at least one; based on the recognition result of the vascular stenosis corresponding to each angiography image, and according to a preset key frame selection strategy, selecting a single key frame from the angiography sequence, wherein the method comprises the following steps: selecting a candidate sequence from the angiography sequence according to the key frame selection strategy based on the vascular stenosis recognition result corresponding to each angiography image; and processing the obtained at least one candidate sequence, and selecting a single key frame from angiography images respectively contained in the at least one candidate sequence according to a processing result. Wherein, the explanation of the same or corresponding terms as the above embodiments is not repeated herein.

Referring to fig. 5, the method of this embodiment may specifically include the following steps:

s210, acquiring an angiography sequence, and identifying vascular stenosis of each angiography image in the angiography sequence.

S220, selecting a candidate sequence from the angiography sequence according to a preset key frame selection strategy based on a vascular stenosis recognition result corresponding to each angiography image respectively.

The number of the key frame selection strategies can be one, two or more, and each key frame selection strategy can be used for selecting the key frame respectively. On the basis, aiming at each key frame selection strategy in at least one key frame selection strategy, a candidate sequence can be selected from an angiography sequence based on a vascular stenosis recognition result respectively corresponding to each angiography image and combined with the key frame selection strategy, and the angiography images in the candidate sequence can be understood as the key frames selected from the angiography sequence based on the key frame selection strategy and are candidates of the finally selected key frames; the number of angiographic images in the candidate sequence may be one, two or more, as the case may be and is not specifically limited herein.

S230, processing the obtained at least one candidate sequence, and selecting a single key frame from angiography images respectively contained in the at least one candidate sequence according to a processing result.

Wherein, after the candidate sequences are respectively selected based on each key frame selection strategy, at least one candidate sequence can be obtained. On the basis, at least one candidate sequence is further processed, and a single key frame is selected from angiographic images respectively contained in the at least one candidate sequence according to the obtained processing result. For example, one or more repeated images existing in at least one candidate sequence may be selected from angiographic images respectively included in at least one candidate sequence, and then a single key frame may be selected from the one or more repeated images; or respectively assigning weights to at least one candidate sequence, and selecting a single key frame from angiographic images respectively contained in at least one candidate sequence according to the weights respectively corresponding to the at least one candidate sequence; etc., and are not particularly limited herein.

According to the technical scheme, candidate sequences are respectively selected through each key frame selection strategy, then, a single key frame is selected from angiography images respectively contained in at least one candidate sequence through processing the selected at least one candidate sequence, and the selected key frame is matched with each key frame selection strategy, so that the accuracy of key frame selection is improved.

Fig. 6 is a flow chart of yet another angiographic image key frame selection method provided in an embodiment of the invention. The present embodiment is optimized based on the above technical solutions. In this embodiment, optionally, according to the processing result, selecting a single keyframe from angiographic images respectively included in at least one candidate sequence may include: selecting at least one candidate frame from angiographic images respectively contained in at least one candidate sequence according to the obtained processing result; aiming at each candidate frame in at least one candidate frame, obtaining the vessel diameter of the candidate frame at the vessel stenosis according to the vessel stenosis identification result corresponding to the candidate frame; and taking the candidate frame with the smallest vessel diameter in the at least one candidate frame as a key frame. The same or corresponding terms as those of the above embodiments are not repeated herein.

Referring to fig. 6, the method of this embodiment may specifically include the following steps:

s310, acquiring an angiography sequence, and identifying vascular stenosis of each angiography image in the angiography sequence.

S320, selecting a candidate sequence from the angiography sequence according to a preset key frame selection strategy based on a vascular stenosis recognition result corresponding to each angiography image.

S330, processing the obtained at least one candidate sequence, and selecting at least one candidate frame from angiography images respectively contained in the at least one candidate sequence according to the obtained processing result.

The candidate frames may be understood as angiographic images that are more suitable for being used as key frames in angiographic images included in at least one candidate sequence, and the number of candidate frames may be one, two or more, which is relevant to practical situations and is not specifically limited herein.

And selecting at least one candidate frame from angiographic images respectively contained in the at least one candidate sequence according to the processing result of the at least one candidate sequence. In combination with the application scenario possibly related to the embodiment of the present invention, optionally, in the case that the key frame selection policy is determined according to the key frame requirement, angiographic images which are more matched with the key frame requirement in angiographic images respectively included in at least one candidate sequence may be obtained according to the processing result, and the obtained angiographic images are used as candidate frames.

S340, obtaining the vessel diameter of the candidate frame at the vessel stenosis position according to the vessel stenosis recognition result corresponding to the candidate frame for each candidate frame in the at least one candidate frame.

According to the related medical knowledge, in each angiography image in the angiography sequence, a key frame for realizing detection of the vascular stenosis is narrower at the vascular stenosis, and the vascular stenosis condition can be determined by the vascular diameter. For this purpose, for each candidate frame of the at least one candidate frame, a vessel diameter of the candidate frame at the vessel stenosis may be obtained based on the vessel stenosis recognition result of the candidate frame.

S350, taking the candidate frame with the smallest blood vessel diameter in at least one candidate frame as a key frame.

After obtaining the vessel diameters respectively corresponding to at least one candidate frame, the minimum diameter with the smallest value in all the vessel diameters can be determined, and then the candidate frame corresponding to the minimum diameter is used as a key frame.

According to the technical scheme provided by the embodiment of the invention, at least one candidate frame which is more suitable for being applied as a key frame is selected from angiography images respectively contained in at least one candidate sequence, and then the candidate frame with the smallest vessel diameter at the vessel stenosis position in the at least one candidate frame is used as the key frame, so that the selected key frame is narrowest at the vessel stenosis position, and the accuracy of vessel stenosis detection can be further improved.

An optional technical solution, according to a recognition result of a vascular stenosis corresponding to a candidate frame, obtains a vascular diameter of the candidate frame at the vascular stenosis, including:

The center line is understood as the center line of the candidate frame in the blood vessel included in the blood vessel stenosis position, and the center line of the candidate frame in the blood vessel stenosis position is determined according to the blood vessel stenosis recognition result of the candidate frame. In practical applications, alternatively, the center line may be determined by a plurality of methods, for example, a vascular stenosis region may be selected from the candidate frames according to a result of identifying the vascular stenosis, and then the center line selected from the blood vessels included in the vascular stenosis region may be used as the center line of the candidate frames at the vascular stenosis; then, for example, a central line can be selected from the blood vessels contained in the candidate frame, and then the selected central line is intercepted according to the recognition result of the blood vessel stenosis to obtain the central line of the candidate frame at the blood vessel stenosis; of course, the center line may be determined by other methods, and is not particularly limited herein. On this basis, optionally, a centerline may be selected from the blood vessel by using an artificial intelligence (Artificial Intelligence, AI) -based training acquisition method, a manual calibration method, a topology refinement method, a distance transformation method, or the like, which is not particularly limited herein.

After obtaining the centerline of the candidate frame at the vessel stenosis, a perpendicular to the centerline is obtained, and then two intersections of the perpendicular on the vessel boundary at the vessel stenosis are obtained. On this basis, in connection with the application scenario possibly related to the embodiment of the present invention, alternatively, a direction perpendicular to the direction in which the center line is located may be taken as a perpendicular direction, and then a straight line including the target point in the candidate frame and located in the perpendicular direction may be taken as a perpendicular line. On the basis, optionally, the target point may be a pixel point on the central line that meets a preset condition, or may be a pixel point randomly selected from the central line, which is not specifically limited herein; the number of target points may be one, two or more, which may be set according to the actual situation, and is not specifically limited herein.

For example, referring to the vascular stenosis region shown in fig. 7, a center line 1 is extracted from the blood vessel included in the vascular stenosis region, and then a perpendicular line 2 to the center line 1 is calculated, thereby obtaining two intersection points 3 (i.e., pixel points circled by dotted lines in fig. 7) of the perpendicular line 2 on the boundary of the blood vessel at the vascular stenosis.

Further, the distance between the two intersections is used as the vessel diameter of the candidate frame at the vessel stenosis. On this basis, optionally, in the case that two or more target points exist, two or more perpendicular lines can be obtained, and then the diameters of blood vessels corresponding to each perpendicular line can be calculated, and then the diameter of the blood vessel with the smallest numerical value in the direct blood vessels is used as the diameter of the blood vessel corresponding to the candidate frame.

According to the technical scheme, the accuracy of determining the blood vessel diameter is improved through pixel-level processing.

Fig. 8 is a flow chart of yet another angiographic image key frame selection method provided in an embodiment of the invention. The present embodiment is optimized based on the above technical solutions. In this embodiment, optionally, processing the obtained at least one candidate sequence, and selecting, according to a processing result, a single keyframe from angiographic images respectively included in the at least one candidate sequence may include: the method comprises the steps of obtaining weights corresponding to at least one obtained candidate sequence respectively, and carrying out assignment processing on the at least one candidate sequence according to the weights corresponding to the at least one candidate sequence respectively; and selecting a single key frame from angiography images respectively contained in at least one candidate sequence according to the obtained assignment processing result. Wherein, the explanation of the same or corresponding terms as the above embodiments is not repeated herein.

Referring to fig. 8, the method of this embodiment may specifically include the following steps:

s410, acquiring an angiography sequence, and identifying vascular stenosis of each angiography image in the angiography sequence.

S420, selecting a candidate sequence from the angiography sequence according to a preset key frame selection strategy based on a vascular stenosis recognition result corresponding to each angiography image respectively.

S430, obtaining weights respectively corresponding to the obtained at least one candidate sequence, and carrying out assignment processing on the at least one candidate sequence according to the weights respectively corresponding to the at least one candidate sequence.

For each candidate sequence in at least one candidate sequence, a weight endowed to the candidate sequence is obtained, and in practical application, optionally, the weight can be determined according to a key frame selection strategy corresponding to the candidate sequence, further can be determined according to a key frame requirement matched with the key frame selection strategy, for example, can be determined according to the duty ratio of the key frame requirement in all key frame requirements, and the duty ratio can characterize the importance of the key frame requirement for vascular stenosis detection to a certain extent; still alternatively, the sum of weights respectively corresponding to the at least one candidate sequence may be 1.

Further, assignment processing is performed on the candidate sequence according to the weight corresponding to the candidate sequence, for example, each angiographic image in the candidate sequence can be respectively assigned with a numerical value associated with the weight.

S440, selecting a single key frame from angiography images respectively contained in at least one candidate sequence according to the obtained assignment processing result.

After the assignment results corresponding to the at least one candidate sequence are obtained, a single key frame can be selected from angiographic images respectively contained in the at least one candidate sequence according to the assignment results.

According to the technical scheme provided by the embodiment of the invention, the key frames can be selected according to the obtained assignment processing results by acquiring the weights respectively corresponding to at least one candidate sequence and carrying out assignment processing on the corresponding candidate sequences according to the weights, and compared with other angiographic images in an angiographic sequence, the characteristics of the selected key frames are more suitable for detecting vascular stenosis, so that the accuracy of vascular stenosis detection is ensured.

An optional technical solution, performing assignment processing on at least one candidate sequence according to weights corresponding to the at least one candidate sequence respectively, includes:

For example, for a certain candidate sequence K1 in at least one candidate sequence, assuming that the candidate sequence K1 includes an angiographic image a, an angiographic image b and an angiographic image c, weights α corresponding to the candidate sequence K1 may be assigned to the angiographic image a, the angiographic image b and the angiographic image c, respectively, and then the assignment processing results of the angiographic image a, the angiographic image b and the angiographic image c are all α.

According to the technical scheme, the accuracy of key frame selection is improved through the assignment processing of the image level.

Wherein for all angiographic images contained in at least one candidate sequence, one or some of the angiographic images may be present in more than two (i.e., two or more) candidate sequences, such angiographic images may be referred to as duplicate images. Illustratively, it is assumed that at least one candidate sequence comprises candidate sequence K1, candidate sequence K2 and candidate sequence K3, wherein candidate sequence K1 comprises angiographic image a, angiographic image b and angiographic image c, candidate sequence K2 comprises angiographic image c and angiographic image d, and candidate sequence K3 comprises angiographic image e, wherein all angiographic images may comprise angiographic image a, angiographic image b, angiographic image c, angiographic image d and angiographic image e, and the repeated image is angiographic image c.

In the case where the assignment process is performed for each angiographic image in the candidate sequence, two or more assignment process results corresponding to the repeated image may be combined, and the combined result may be used as the assignment process result for the repeated image. Illustratively, assuming that the candidate sequence K1 corresponds to the weight α, the candidate sequence K2 corresponds to the weight β, and the candidate sequence K3 corresponds to the weight γ, the assignment processing results of the angiographic image a and the angiographic image b are α, the assignment processing result of the angiographic image d is β, the assignment processing result of the angiographic image e is γ, and the assignment processing result of the angiographic image c (i.e., the repeated image) is α+β. Further, a single key frame is selected from all angiographic images according to the assigned processing results respectively corresponding to all angiographic images.

According to the technical scheme, the assignment processing results of the repeated images are combined, so that more accurate assignment processing results of each angiographic image in all angiographic images contained in at least one candidate sequence can be obtained, and further, key frames capable of detecting vascular stenosis more accurately can be selected based on the more accurate assignment processing results.

In order to better understand the above-described respective technical solutions as a whole, an exemplary description thereof is given below in conjunction with specific examples. Exemplary, as shown in fig. 9, candidate sequence K1 is selected from the angiographic sequence according to a key frame selection policy corresponding to the first type of requirement, candidate sequence K2 is selected from the angiographic sequence according to a key frame selection policy corresponding to the second type of requirement, and candidate sequence K3 is selected from the angiographic sequence according to a key frame selection policy corresponding to the third type of requirement. Further, according to weights corresponding to the candidate sequence K1, the candidate sequence K2 and the candidate sequence K3, assignment processing is performed on angiographic images in the three candidate sequences respectively, and assignment processing results of repeated images in the three candidate sequences are combined, so that the score of each angiographic image in all angiographic images contained in the three candidate sequences can be obtained. Further, a candidate sequence K4 with the score larger than a preset score threshold value is selected from all angiographic images, and an angiographic image with the direct smallest blood vessel is selected from the candidate sequence K4 to serve as a key frame.

In the above example, keyframes may be selected that meet the first, second, and third types of requirements and are narrowest at a stenosis of a blood vessel.

Fig. 10 is a block diagram of an angiographic image key frame selection device according to an embodiment of the invention, which may be used to perform the angiographic image key frame selection method according to any of the above embodiments. The device and the angiographic image key frame selection method of the above embodiments belong to the same inventive concept, and reference may be made to the embodiment of the angiographic image key frame selection method for details which are not described in detail in the embodiment of the angiographic image key frame selection device. Referring to fig. 10, the apparatus may specifically include: a vascular stenosis identification module 510 and a keyframe selection module 520.

The angiostenosis recognition module 510 is configured to acquire an angiography sequence, and perform angiostenosis recognition on each angiography image in the angiography sequence;

the key frame selection module 520 is configured to select a single key frame from the angiographic sequence based on the respective recognition result of the angiostenosis of each angiographic image and according to a preset key frame selection policy.

Optionally, the number of key frame selection policies is at least one; the key frame selection module 520 includes:

the candidate sequence selection sub-module is used for selecting a strategy for each key frame in at least one key frame selection strategy, selecting a candidate sequence from the angiography sequence based on a vascular stenosis recognition result respectively corresponding to each angiography image and according to the key frame selection strategy;

and the key frame selection sub-module is used for processing the obtained at least one candidate sequence and selecting a single key frame from angiography images respectively contained in the at least one candidate sequence according to a processing result.

On this basis, an optional key frame selection submodule includes:

a candidate frame selecting unit, configured to select at least one candidate frame from angiographic images respectively included in at least one candidate sequence according to the obtained processing result;

a blood vessel diameter obtaining unit, configured to obtain, for each candidate frame in at least one candidate frame, a blood vessel diameter of the candidate frame at a blood vessel stenosis according to a blood vessel stenosis identification result corresponding to the candidate frame;

and the key frame first selecting unit is used for taking the candidate frame with the smallest blood vessel diameter in the at least one candidate frame as the key frame.

On this basis, optionally, the vessel diameter obtaining unit includes:

the intersection point obtaining subunit is used for determining the central line of the candidate frame at the vascular stenosis according to the vascular stenosis identification result corresponding to the candidate frame to obtain two intersection points of the perpendicular line of the central line on the vascular boundary of the vascular stenosis;

the vessel diameter obtaining subunit is used for taking the distance between two intersection points as the vessel diameter of the candidate frame at the narrow part of the vessel.

Alternatively, the key frame selecting sub-module includes:

the assignment processing unit is used for acquiring weights respectively corresponding to the obtained at least one candidate sequence and carrying out assignment processing on the at least one candidate sequence according to the weights respectively corresponding to the at least one candidate sequence;

and the key frame second selecting unit is used for selecting a single key frame from angiography images respectively contained in at least one candidate sequence according to the obtained assignment processing result.

On the basis, optionally, the assignment processing unit includes:

and the assignment processing subunit is used for respectively carrying out assignment processing on each angiographic image in the candidate sequence according to the weight corresponding to the candidate sequence aiming at each candidate sequence in the at least one candidate sequence.

Optionally, in the case that assignment processing is performed on each angiographic image in the candidate sequence, the key frame second selection unit is specifically configured to:

Optionally, the angiography image key frame selecting device further includes:

and the vessel segmentation image obtaining module is used for carrying out vessel segmentation on the angiographic images after aiming at each angiographic image in the angiographic sequence to obtain vessel segmentation images.

On this basis, optionally, the key frame selection module 520 is specifically configured to:

According to the angiography image key frame selection device provided by the embodiment of the invention, an angiography sequence is acquired through the angiostenosis recognition module, and angiostenosis recognition is performed on angiography images aiming at each angiography image in the angiography sequence; then, through a key frame selection module, a single key frame is selected from an angiography sequence based on a vascular stenosis identification result corresponding to each angiography image respectively and according to a preset key frame selection strategy. The device can automatically select the key frames from the angiography sequence, thereby reducing the workload of doctors.

The angiography image key frame selecting device provided by the embodiment of the invention can execute the angiography image key frame selecting method provided by any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the executing method.

It should be noted that, in the embodiment of the angiographic image key frame selection apparatus, each unit and module included are only divided according to the functional logic, but are not limited to the above-mentioned division, so long as the corresponding functions can be implemented; in addition, the specific names of the functional units are also only for distinguishing from each other, and are not used to limit the protection scope of the present invention.

Fig. 11 shows a schematic diagram of the structure of an electronic device 10 that may be used to implement an embodiment of the invention. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. Electronic equipment may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the inventions described and/or claimed herein.

As shown in fig. 11, the electronic device 10 includes at least one processor 11, and a memory such as a Read Only Memory (ROM) 12, a Random Access Memory (RAM) 13, etc., communicatively connected to the at least one processor 11, wherein the memory stores a computer program executable by the at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the Read Only Memory (ROM) 12 or the computer program loaded from the storage unit 18 into the Random Access Memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 may also be stored. The processor 11, the ROM 12 and the RAM 13 are connected to each other via a bus 14. An input/output (I/O) interface 15 is also connected to bus 14.

Various components in the electronic device 10 are connected to the I/O interface 15, including: an input unit 16 such as a keyboard, a mouse, etc.; an output unit 17 such as various types of displays, speakers, and the like; a storage unit 18 such as a magnetic disk, an optical disk, or the like; and a communication unit 19 such as a network card, modem, wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information/data with other devices via a computer network, such as the internet, and/or various telecommunication networks.

The processor 11 may be a variety of general and/or special purpose processing components having processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various processors running machine learning model algorithms, digital Signal Processors (DSPs), and any suitable processor, controller, microcontroller, etc. The processor 11 performs the various methods and processes described above, such as the angiographic image key frame selection method.

In some embodiments, the angiographic image key frame selection method may be implemented as a computer program tangibly embodied on a computer-readable storage medium, such as the storage unit 18. In some embodiments, part or all of the computer program may be loaded and/or installed onto the electronic device 10 via the ROM 12 and/or the communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the angiographic image key-frame selection method described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to perform the angiographic image keyframe selection method in any other suitable way (e.g. by means of firmware).

Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuit systems, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems On Chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.

A computer program for carrying out methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the computer programs, when executed by the processor, cause the functions/acts specified in the flowchart and/or block diagram block or blocks to be implemented. The computer program may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.

In the context of the present invention, a computer-readable storage medium may be a tangible medium that can contain, or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer readable storage medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. Alternatively, the computer readable storage medium may be a machine readable signal medium. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

To provide for interaction with a user, the systems and techniques described here can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) through which a user can provide input to the electronic device. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.

The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), blockchain networks, and the internet.

The computing system may include clients and servers. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also called a cloud computing server or a cloud host, and is a host product in a cloud computing service system, so that the defects of high management difficulty and weak service expansibility in the traditional physical hosts and VPS service are overcome.

It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps described in the present invention may be performed in parallel, sequentially, or in a different order, so long as the desired results of the technical solution of the present invention are achieved, and the present invention is not limited herein.

The above embodiments do not limit the scope of the present invention. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of the present invention.

Claims

1. A method for selecting a keyframe of an angiographic image, comprising:

acquiring an angiography sequence, and identifying vascular stenosis of each angiography image in the angiography sequence;

2. The method of claim 1, wherein the number of keyframe selection policies is at least one;

the step of selecting a single key frame from the angiography sequence based on the respective corresponding vascular stenosis recognition results of each angiography image and according to a preset key frame selection strategy comprises the following steps:

selecting a candidate sequence from the angiography sequence according to each key frame selection strategy in the at least one key frame selection strategy based on the vascular stenosis recognition result respectively corresponding to each angiography image;

3. The method according to claim 2, wherein selecting a single keyframe from angiographic images respectively contained in the at least one candidate sequence according to the processing result comprises:

selecting at least one candidate frame from angiographic images respectively contained in the at least one candidate sequence according to the obtained processing result;

For each candidate frame in the at least one candidate frame, obtaining the vessel diameter of the candidate frame at the vessel stenosis according to the vessel stenosis identification result corresponding to the candidate frame;

4. A method according to claim 3, wherein the obtaining the vessel diameter of the candidate frame at the vessel stenosis according to the vessel stenosis identification result corresponding to the candidate frame comprises:

and taking the distance between the two intersection points as the vessel diameter of the candidate frame at the vessel stenosis.

5. The method according to claim 2, wherein the processing the obtained at least one candidate sequence, and selecting a single key frame from angiographic images respectively contained in the at least one candidate sequence according to the processing result, includes:

And selecting a single key frame from angiography images respectively contained in the at least one candidate sequence according to the obtained assignment processing result.

6. The method according to claim 5, wherein assigning the at least one candidate sequence according to the weights respectively corresponding to the at least one candidate sequence comprises:

7. The method according to claim 6, wherein, in the case of performing assignment processing on each angiographic image in the candidate sequence, selecting a single keyframe from angiographic images respectively included in the at least one candidate sequence according to the obtained assignment processing result includes:

combining the two or more assignment processing results corresponding to the repeated images in the repeated images existing in the two or more candidate sequences in all angiographic images contained in the at least one candidate sequence, and taking the combined result as the assignment processing result of the repeated images;

8. The method of claim 1, wherein the key frame selection policy is determined based on key frame requirements, wherein the key frame requirements characterize requirements to be met by a key frame for vascular stenosis detection.

9. The method of claim 8, wherein the class number of keyframe requirements includes at least one class, each class of keyframe requirements corresponding to a different keyframe selection policy.

10. The method of claim 9, wherein the at least one type of keyframe requirements includes at least one of a first type of requirements, a second type of requirements, and a third type of requirements;

wherein the first class of requirements includes at least one of intravascular contrast agent filling, vessel in a relaxed state, and vessel non-foreshortening at a vessel stenosis;

the third class of requirements includes at least one of no artifact at the vessel stenosis and a clear vessel boundary at the vessel stenosis.

11. The method of claim 1, wherein the step of, for each angiographic image in the angiographic sequence, further comprises:

And carrying out blood vessel segmentation on the angiography image to obtain a blood vessel segmentation image.

12. The method according to claim 11, wherein selecting a single keyframe from the angiographic sequence based on the respective angiostenosis recognition result of each angiographic image and according to a preset keyframe selection policy comprises:

13. An angiographic image key frame selection device, comprising:

the vascular stenosis recognition module is used for acquiring an angiography sequence and recognizing vascular stenosis of each angiography image in the angiography sequence;

and the key frame selection module is used for selecting a single key frame from the angiography sequence based on the vascular stenosis identification result corresponding to each angiography image and according to a preset key frame selection strategy.

14. An electronic device, comprising:

At least one processor; and

a memory communicatively coupled to the at least one processor; wherein,,

the memory stores a computer program executable by the at least one processor to cause the at least one processor to perform the angiographic image keyframe selection method of any one of claims 1-12.

15. A computer readable storage medium storing computer instructions for causing a processor to perform the angiographic image key-frame selection method according to any one of claims 1-12.