CN114418977B - Method and device for quantitative analysis of coronary angiography based on angiography video - Google Patents
Method and device for quantitative analysis of coronary angiography based on angiography video Download PDFInfo
- Publication number
- CN114418977B CN114418977B CN202210018415.6A CN202210018415A CN114418977B CN 114418977 B CN114418977 B CN 114418977B CN 202210018415 A CN202210018415 A CN 202210018415A CN 114418977 B CN114418977 B CN 114418977B
- Authority
- CN
- China
- Prior art keywords
- image
- pixel point
- blood vessel
- center line
- video
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30101—Blood vessel; Artery; Vein; Vascular
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30172—Centreline of tubular or elongated structure
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computational Linguistics (AREA)
- Radiology & Medical Imaging (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Geometry (AREA)
- Medical Informatics (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Quality & Reliability (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
本发明实施例涉及一种基于血管造影视频进行冠脉造影定量分析的方法和装置，所述方法包括：获取血管造影视频；进行视频帧图像提取生成第一图像序列；基于图像目标检测与语义分割模型对第一图像序列中各个第一图像进行狭窄段血管的目标检测和语义分割处理从而在每个第一图像上得到一个或多个用于标记狭窄段血管的目标检测框以及在每个目标检测框中的一段狭窄段血管掩膜图像；对第一图像序列进行图像优选处理得到指定数量的优选图像；在各个优选图像上对每个目标检测框内的狭窄段血管掩膜图像进行冠脉造影定量分析生成对应的血管狭窄率。通过本发明可摆脱传统做法中对人工经验的过度依赖，提高图像提取准确度和狭窄率计算精度。
The embodiment of the present invention relates to a method and device for quantitative analysis of coronary angiography based on angiography video, the method comprising: obtaining angiography video; extracting video frame images to generate a first image sequence; performing target detection and semantic segmentation processing on each first image in the first image sequence for stenotic segment vessels, based on an image target detection and semantic segmentation model, thereby obtaining one or more target detection frames for marking stenotic segment vessels and a stenotic segment vessel mask image in each target detection frame on each first image; performing image optimization processing on the first image sequence to obtain a specified number of preferred images; performing coronary angiography quantitative analysis on the stenotic segment vessel mask image in each target detection frame on each preferred image to generate a corresponding vessel stenosis rate. The present invention can get rid of the excessive reliance on manual experience in traditional practices, and improve the accuracy of image extraction and the accuracy of stenosis rate calculation.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a method and a device for quantitative analysis of coronary angiography based on angiography video.
Background
Coronary stenosis can lead to insufficient blood supply to the heart, causing dysfunction and/or lesions in the heart muscle. The angiography technology is based on the principle that X-rays cannot penetrate through a developer, the developer is injected into a blood vessel of a detection object, and the developer under the X-rays is subjected to image shooting in the process of passing through the blood vessel so as to output angiography video. When detecting the stenosis of the coronary artery, an angiography video of the coronary vessel is obtained based on an angiography technology in the conventional case, and then a doctor screens out a video image with the vessel at the stenosis from the angiography video according to personal experience to perform quantitative analysis of the vessel stenosis rate, which is also called qualitative comparison analysis (Qualitative Comparative Analysis, QCA), so as to calculate the corresponding vessel stenosis rate. The operation mode is too dependent on human factors, such as personnel experience, human eye recognition capability and the like, and the problems of inaccurate video image extraction, insufficient accuracy of stenosis rate calculation and the like are very easy to occur.
Disclosure of Invention
The invention aims at overcoming the defects of the prior art and provides a method, a device, electronic equipment and a computer readable storage medium for quantitative analysis of coronary angiography based on angiography video, which are used for carrying out video interception and video frame image extraction processing on the angiography video of coronary vessels, carrying out target detection and semantic segmentation processing on extracted image sequences of narrow-section vessels based on an image target detection and semantic segmentation model, carrying out optimization on extracted images based on the confidence of target identification, and carrying out quantitative analysis of coronary angiography on each narrow-section vessel on the optimized images to generate corresponding vascular stenosis rate. The invention can get rid of excessive dependence on manual experience in the traditional method and improve the image extraction accuracy and the stenosis rate calculation accuracy.
To achieve the above object, according to a first aspect of the present invention, there is provided a method for quantitative analysis of coronary angiography based on angiography video, the method comprising:
acquiring angiography video of coronary angiography;
extracting video frame images of the angiography video to generate a corresponding first image sequence;
Performing target detection and semantic segmentation processing on each first image in the first image sequence based on a preset image target detection and semantic segmentation model, so as to obtain one or more target detection frames for marking the narrow blood vessel and a narrow blood vessel mask image in each target detection frame on each first image, wherein each target detection frame corresponds to one detection frame confidence;
Performing image optimization processing on the first image sequence according to the confidence coefficient of the detection frame to obtain an appointed number of optimized images;
and carrying out coronary angiography quantitative analysis on the mask images of the stenosis blood vessel in each target detection frame on each preferable image to generate a corresponding blood vessel stenosis rate.
Preferably, the video frame image extraction of the angiography video generates a corresponding first image sequence, which specifically includes:
Video interception is carried out on the angiography video, and video content of a contrast agent filling coronary artery stage is reserved to generate a corresponding intercepted angiography video;
performing video frame image extraction processing on the intercepted contrast video according to time sequence to generate a corresponding video frame image sequence, and counting the number of video frame images of the video frame image sequence to generate a corresponding first total number;
When the first total number does not exceed a preset total number threshold value of images, taking each video frame image as a corresponding first image, and sequencing all the obtained first images according to time sequence to generate the first image sequence;
When the first total number exceeds the total number threshold of the images, extracting the video frame images with all odd indexes from the video frame image sequence to serve as the corresponding first images, or extracting the video frame images with all even indexes from the video frame image sequence to serve as the corresponding first images, and sorting all the obtained first images according to time sequence to generate the first image sequence.
Preferably, the image target detection and semantic segmentation model comprises a Mask R-CNN model, and when the image target detection and semantic segmentation model is specifically the Mask R-CNN model, a residual network ResNet is adopted as a characteristic extraction backbone network.
Preferably, the performing image optimization processing on the first image sequence according to the confidence coefficient of the detection frame to obtain a specified number of preferred images specifically includes:
in the first image sequence, carrying out mean value calculation on the detection frame confidence degrees of all the target detection frames on each first image to generate corresponding first image average confidence degrees;
And sequencing all the first images according to the sequence from the larger average confidence level to the smaller average confidence level of the corresponding first images, and taking the appointed number of first images sequenced in front as the preferable images.
Preferably, the performing coronary angiography quantitative analysis on the mask image of the stenosis segment in each target detection frame on each preferred image to generate a corresponding vascular stenosis rate specifically includes:
Traversing each target detection frame on the current preferred image, and marking the currently traversed target detection frame as a current target detection frame;
carrying out vessel edge and vessel center line identification on the narrow-section vessel mask image in the current target detection frame to generate a corresponding first vessel edge and a first center line, wherein the first center line comprises a plurality of center line pixel points P i, the first center line pixel point P 1 is the closest point to a coronary artery inlet in the blood flow direction, the last center line pixel point P N is the farthest point from the coronary artery inlet in the blood flow direction, i is more than or equal to 1 and less than or equal to N, and N is the total number of the center line pixel points of the first center line;
Analyzing the blood vessel diameter length corresponding to each central line pixel point P i on the first central line according to the first blood vessel edge to generate a corresponding first blood vessel diameter d i;
According to the linear change relation of the blood vessels from the central line pixel point P 1 to the central line pixel point P N and the first blood vessel diameter d i of each central line pixel point P i, analyzing the corresponding stenosis rate of each central line pixel point P i to generate a corresponding first stenosis rate r i;
and selecting a maximum value from all the obtained first stenosis rates r i as the vessel stenosis rate corresponding to the stenosis segment vessel mask image in the current target detection frame.
Further, according to the first blood vessel edge, analyzing the blood vessel diameter length corresponding to each central line pixel point P i on the first central line to generate a corresponding first blood vessel diameter d i, which specifically includes:
The center line pixel point P i and the adjacent eight-domain pixel points thereof are respectively marked as a first straight line, a second straight line, a third straight line and a fourth straight line by making four straight lines according to the direction relation of the center line pixel point P i, wherein the first straight line passes through the upper left adjacent pixel point of the center line pixel point P i, the center line pixel point P i and the lower right adjacent pixel point of the center line pixel point P i, the second straight line passes through the upper adjacent pixel point of the center line pixel point P i, the center line pixel point P i and the lower adjacent pixel point of the center line pixel point P i, the third straight line passes through the upper right adjacent pixel point of the center line pixel point P i, the center line pixel point P i and the lower left adjacent pixel point of the center line pixel point P i, and the third straight line passes through the right adjacent pixel point of the center line pixel point P i, the center line pixel point P i and the left adjacent pixel point P i;
The first, second, third and fourth straight lines are respectively marked as corresponding first, second, third and fourth line segments, the line segment lengths of the first, second, third and fourth line segments are calculated to generate corresponding first, second, third and fourth line segment lengths, and the minimum value is selected from the first, second, third and fourth line segment lengths to be used as the first blood vessel diameter d i corresponding to the central line pixel point P i.
Further, according to the linear change relationship between the central line pixel point P 1 and the blood vessel of the central line pixel point P N and the first blood vessel diameter d i of each central line pixel point P i, the analyzing the corresponding stenosis rate of each central line pixel point P i to generate a corresponding first stenosis rate r i specifically includes:
Constructing a linear function f (i), f (i) =d 1+k*(i-1),k＝(dN-d1)/(N-1) reflecting the linear variation relation of the blood vessels from the central line pixel point P 1 to the central line pixel point P N according to the first blood vessel diameter d 1 and the first blood vessel diameter d N;
Calculating the length of the linear variation diameter corresponding to each central line pixel point P i according to the linear variation relation function f (i) to generate a corresponding first reference diameter d ' i;
And calculating the first stenosis rate r i,ri＝1-di/d' i corresponding to each central line pixel point P i according to the first blood vessel diameter d i and the first reference diameter d ' i.
A second aspect of the embodiment of the present invention provides an apparatus for implementing the method described in the first aspect, where the apparatus includes an acquisition module, an image preprocessing module, a stenosis blood vessel processing module, an image optimization module, and a quantitative analysis module;
The acquisition module is used for acquiring angiography video of coronary angiography;
The image preprocessing module is used for extracting video frame images of the angiography video to generate a corresponding first image sequence;
The narrow-section blood vessel processing module is used for carrying out target detection and semantic segmentation processing on narrow-section blood vessels on each first image in the first image sequence based on a preset image target detection and semantic segmentation model, so that one or more target detection frames for marking the narrow-section blood vessels and a narrow-section blood vessel mask image in each target detection frame are obtained on each first image;
the image optimization module is used for performing image optimization processing on the first image sequence according to the confidence coefficient of the detection frame to obtain an appointed number of optimized images;
And the quantitative analysis module is used for carrying out coronary angiography quantitative analysis on the mask images of the stenosis blood vessel in each target detection frame on each preferable image to generate a corresponding blood vessel stenosis rate.
A third aspect of the embodiment of the invention provides an electronic device, comprising a memory, a processor and a transceiver;
The processor is configured to couple to the memory, and read and execute the instructions in the memory, so as to implement the method steps described in the first aspect;
the transceiver is coupled to the processor and is controlled by the processor to transmit and receive messages.
A fourth aspect of the embodiments of the present invention provides a computer-readable storage medium storing computer instructions that, when executed by a computer, cause the computer to perform the instructions of the method of the first aspect.
The embodiment of the invention provides a method, a device, electronic equipment and a computer-readable storage medium for quantitative analysis of coronary angiography based on angiography video, which are used for carrying out video interception and video frame image extraction processing on the angiography video of coronary vessels, carrying out target detection and semantic segmentation processing on narrow-section vessels on an extracted image sequence based on an image target detection and semantic segmentation model, optimizing an extracted image based on the confidence of target identification, and carrying out quantitative analysis of coronary angiography on each narrow-section vessel on the optimized image to generate corresponding vascular stenosis rate. The invention gets rid of the excessive dependence on manual experience in the traditional method and improves the image extraction accuracy and the stenosis rate calculation accuracy.
Drawings
FIG. 1 is a schematic diagram of a method for quantitative analysis of coronary angiography based on angiography video according to a first embodiment of the present invention;
Fig. 2 is a block diagram of a device for quantitative analysis of coronary angiography based on angiography video according to a second embodiment of the present invention;
Fig. 3 is a schematic structural diagram of an electronic device according to a third embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in further detail below with reference to the accompanying drawings, and 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 invention without making any inventive effort, are intended to be within the scope of the invention.
As shown in fig. 1, which is a schematic diagram of a method for performing quantitative analysis of coronary angiography based on angiography video, the method mainly includes the following steps:
step 1, acquiring angiography video of coronary angiography.
Here, the coronary angiography is a process of injecting a contrast medium into a blood vessel to be detected and imaging the contrast medium under X-rays through the coronary blood vessel, and the angiographic video is video data obtained by the imaging.
Step 2, extracting video frame images of angiography video to generate a corresponding first image sequence;
Step 21, video interception is carried out on angiography video, and video content of a contrast agent filling coronary artery stage is reserved to generate a corresponding intercepted angiography video;
Here, since the angiographic video includes the whole process video content from the injection of the contrast agent into the blood vessel to the gradual filling of the whole coronary artery to the gradual dissipation, the embodiment of the invention focuses on the video content after the contrast agent reaches the coronary artery, so that the angiographic video is required to be intercepted in advance for improving the data analysis efficiency, wherein one of the modes is that a relative time threshold can be set according to the implementation experience of the coronary angiography and the video data before the relative time threshold in the angiographic video is intercepted, and the video data after the relative time threshold is reserved as the video content of the contrast agent filling coronary artery stage to generate a corresponding intercepted angiographic video;
step 22, carrying out video frame image extraction processing on the intercepted contrast video according to time sequence to generate a corresponding video frame image sequence, and carrying out statistics on the number of video frame images of the video frame image sequence to generate a corresponding first total number, wherein the video frame image sequence comprises a plurality of video frame images;
Each type of video data defaults to have a corresponding video sampling frame rate parameter, the video frame images of the intercepted contrast video are extracted according to the video sampling frame rate parameter corresponding to the angiography video, each extracted frame image is a video frame image, and the video frame images are sequenced according to time sequence to obtain a video frame image sequence;
And step 23, when the first total number does not exceed the preset total number of images threshold, taking each video frame image as a corresponding first image, and sequencing all the obtained first images according to time sequence to generate a first image sequence, when the first total number exceeds the total number of images threshold, extracting video frame images with all odd indexes from the video frame image sequence as corresponding first images, or extracting video frame images with all even indexes from the video frame image sequence as corresponding first images, and sequencing all the obtained first images according to time sequence to generate the first image sequence.
The method for controlling the image quantity of the video frame image sequence comprises the steps of setting a total image threshold value in advance, identifying whether the total image quantity in the video frame image sequence, namely a first total image quantity, exceeds the threshold value, directly taking each video frame image as a first image if the total image quantity of the video frame image sequence does not exceed the threshold value, sending the first image sequence formed by the first image to the subsequent step for processing if the total image quantity of the video frame image sequence exceeds the threshold value, and carrying out frame extraction reduction on adjacent images if the total image quantity of the video frame image sequence does not exceed the threshold value.
Step 3, performing target detection and semantic segmentation processing on each first image in the first image sequence based on a preset image target detection and semantic segmentation model, so as to obtain one or more target detection frames for marking the narrow-section blood vessel and a narrow-section blood vessel mask image in each target detection frame on each first image;
Each target detection frame corresponds to a detection frame confidence coefficient, the image target detection and semantic segmentation model comprises a Mask R-CNN model, and when the image target detection and semantic segmentation model is specifically the Mask R-CNN model, a residual network ResNet is adopted as a characteristic extraction backbone network.
The image target detection and semantic segmentation model is used for carrying out narrow-segment blood vessel target detection on the input first image, so that one or more target detection frames for marking narrow-segment blood vessels are obtained, and the detection frame confidence of each target detection frame is used for identifying the image in the frame as the credibility of the narrow-segment blood vessel image;
When the image target detection and semantic segmentation model is specifically a Mask R-CNN model, the neural network structure Of the image target detection and semantic segmentation model can refer to articles Mask R-CNN published by authors KAIMING HE, georgia Gkioxari, piotr Doll' ar and Ross Girshick, and the image target detection and semantic segmentation model comprises a feature extraction network layer, a Region candidate network (Region Proposal Network, RPN) layer, a Region alignment (Region Of INTEREST ALIGN, ROI alignment) network layer and a Region HEAD (ROI HEAD) network layer, wherein the feature extraction network layer is connected with the Region candidate network layer, the Region candidate network layer is connected with the Region alignment network layer, the Region alignment network layer is connected with the Region HEAD network layer, the Region HEAD network layer comprises two sub-networks which are a target detection branch network and a target segmentation branch network respectively, and the target detection branch network is used for outputting a target detection frame and a detection frame confidence Of a narrow-segment blood vessel Mask image;
The feature extraction Network layer of the embodiment of the invention is specifically composed of a five-level Residual Network (ResNet) and a corresponding five-level feature pyramid Network (Feature Pyramid Networks, FPN), wherein the region candidate Network layer comprises five-level region candidate networks corresponding to the five-level feature pyramid Network, and when the five-level Residual Network is realized, the embodiment of the invention is realized by using a ResNet-50 Network structure and is used as a backbone Network for feature extraction.
Step 4, performing image optimization processing on the first image sequence according to the confidence coefficient of the detection frame to obtain an appointed number of optimized images;
The method comprises the steps of carrying out mean value calculation on the confidence levels of all target detection frames on each first image in a first image sequence to generate corresponding first image average confidence levels, sequencing all the first images according to the sequence of the corresponding first image average confidence levels from large to small, and taking the appointed number of first images sequenced in front as preferred images.
The higher the average confidence of the first images is, the more obvious the features of the blood vessels of the stenosis on the corresponding first images are, the designated number is set according to specific requirements, for example, the designated number is 3, and the current step can extract 3 first images with the most obvious features of the blood vessels of the stenosis from the first image sequence as preferred images.
Step 5, carrying out coronary angiography quantitative analysis on the mask image of the stenosis section in each target detection frame on each preferable image to generate a corresponding vascular stenosis rate;
Step 51, traversing each target detection frame on the current preferred image, and marking the currently traversed target detection frame as a current target detection frame;
Step 52, performing vessel edge and vessel center line identification on the narrow section vessel mask image in the current target detection frame to generate a corresponding first vessel edge and a first center line;
The first central line comprises a plurality of central line pixel points P i, wherein the first central line pixel point P 1 is the point closest to the coronary artery inlet in the blood flow direction, the last central line pixel point P N is the point farthest from the coronary artery inlet in the blood flow direction, i is more than or equal to 1 and less than or equal to N, and N is the total number of the central line pixel points of the first central line;
The method comprises the steps of carrying out binarization processing on the image content of a current target detection frame to obtain a first binary image, converting pixel values of all pixel points of an original narrow-section vascular mask image on the first binary image into preset foreground pixel values A, converting pixel values of all pixel points except the narrow-section vascular mask image into preset background pixel values B, carrying out point-by-point traversal on each pixel point with the pixel value of the first binary image as the foreground pixel value A, and taking the pixel value of one of eight adjacent pixel points in the current traversal as the background pixel value B, wherein after the traversal is finished, a closed curve obtained by sequentially connecting all the edge points in a clockwise or anticlockwise mode is the blood vessel edge;
The method comprises the steps of performing a first central line extraction process on a narrow-section vascular mask image of a first binary image based on a topology refinement method on the premise of not changing the topological property of the vascular image, wherein the first central line is generated by performing a first central line extraction process on the narrow-section vascular mask image of the first binary image based on the topology refinement method;
When the first central line is obtained, a direction is marked for the central line, a point closest to the coronary artery inlet, namely a narrow segment inlet point, is specially used as a first central line pixel point P 1 of the first central line, and a point farthest from the coronary artery inlet, namely a narrow segment outlet point, is specially used as a last central line pixel point P N of the first central line;
Step 53, according to the edge of the first blood vessel, analyzing the blood vessel diameter length corresponding to each central line pixel point P i on the first central line to generate a corresponding first blood vessel diameter d i;
Step 531, according to the direction relation of the central line pixel point P i and the adjacent eight domain pixel points, the central line pixel point P i is marked as the first, the second, the third and the fourth straight lines respectively;
The first line passing center line pixel point P i, the upper left adjacent pixel point, the center line pixel point P i and the lower right adjacent pixel point of the center line pixel point P i, the second line passing center line pixel point P i, the center line pixel point P i and the lower adjacent pixel point of the center line pixel point P i, the third line passing center line pixel point P i, the upper right adjacent pixel point P i and the lower left adjacent pixel point of the center line pixel point P i, the third line passing center line pixel point P i, the right adjacent pixel point P i and the left adjacent pixel point P i;
Step 532, the line segments of the first, second, third and fourth straight lines intersecting the first vessel edge are respectively marked as corresponding first, second, third and fourth line segments, the line segment lengths of the first, second, third and fourth line segments are calculated to generate corresponding first, second, third and fourth line segment lengths, and the minimum value is selected from the first, second, third and fourth line segment lengths as the first vessel diameter d i corresponding to the central line pixel point P i;
Here, it is first known that the vessel diameter from the over-center line pixel point P i to the first vessel edge should be within the straight line range of all the over-center line pixel points P i, but that the straight line can be actually made only by 4 lines from any one of the over-center line pixel points P i to the first vessel edge, namely, the vessel diameter from the over-center line pixel point P i to the first vessel edge can be only one of the first, second, third and fourth line segments;
Step 54, analyzing the stenosis rate corresponding to each centerline pixel point P i to generate a corresponding first stenosis rate r i according to the vessel linear change relationship from the centerline pixel point P 1 to the centerline pixel point P N and the first vessel diameter d i of each centerline pixel point P i;
Specifically, step 541 is to construct a linear function f (i), f (i) =d 1+k*(i-1),k＝(dN-d1)/(N-1) reflecting the linear variation relationship of the blood vessels from the center line pixel point P 1 to the center line pixel point P N according to the first blood vessel diameter d 1 and the first blood vessel diameter d N;
Step 542, calculating the length of the linear variation diameter corresponding to each centerline pixel point P i according to the linear variation relation function f (i) to generate a corresponding first reference diameter d ' i;
Under the condition that no vascular stenosis mutation occurs, the vascular diameter has a certain linear relation with the distance between the vascular diameter and the coronary artery inlet, and the pipe diameter of the inlet and the outlet of a section of branched blood vessel also has a certain linear relation, so that the normal pipe diameter of any point on the section of blood vessel, namely the first reference diameter d ' i, can be obtained by confirming the linear relation of the inlet and the outlet of the section of blood vessel;
Step 543, calculating a first stenosis rate r i,ri＝1-di/d' i corresponding to each centerline pixel point P i according to the first vessel diameter d i and the first reference diameter d ' i;
Here, if a stenosis mutation occurs at a certain position in a certain segment of blood vessel, after obtaining the stenosis diameter at the position, that is, the first blood vessel diameter d i, the stenosis rate of the blood vessel at the position, that is, the first stenosis rate r i, can be calculated based on the corresponding first reference diameter d ' i;
Step 55, selecting the maximum value from all the obtained first stenosis rates r i as the vessel stenosis rate corresponding to the stenosis segment vessel mask image in the current target detection frame;
here, the embodiment of the invention takes the maximum stenosis rate in a section of blood vessel as the vessel stenosis rate of the section of blood vessel, namely, the mask image of the stenosis section blood vessel in the current target detection frame;
Step 56, the next unprocessed target detection frame is taken as the current target detection frame, and the processing proceeds to step 52 until the vascular stenosis rate of the mask image of the stenosis segment in all the target detection frames on the currently preferred image is confirmed.
Through the steps 1-5, a plurality of preferred images can be extracted from a section of angiographic video, and the vascular stenosis rate of one or more stenosis vessels on each preferred image can be obtained through quantitative analysis of coronary angiography. After the analysis results of the preferred images are obtained, the preferred images with the narrow-section blood vessel target detection frame and the blood vessel stenosis rate can be simultaneously provided for a doctor to serve as parameter data, the preferred images can be subjected to image fusion, the blood vessel stenosis rate of the narrow-section blood vessel at the same position is subjected to mean value calculation, and finally the fused image with the narrow-section blood vessel target detection frame and the blood vessel stenosis rate mean value is provided for the doctor to serve as parameter data.
Fig. 2 is a block diagram of an apparatus for performing quantitative analysis of coronary angiography based on angiography video according to a second embodiment of the present invention, where the apparatus may be a terminal device or a server for implementing a method according to an embodiment of the present invention, or may be an apparatus for implementing a method according to an embodiment of the present invention, which is connected to the terminal device or the server, and for example, the apparatus may be an apparatus or a chip system of the terminal device or the server. As shown in fig. 2, the device comprises an acquisition module 201, an image preprocessing module 202, a stenosis vessel processing module 203, an image optimization module 204 and a quantitative analysis module 205.
The acquisition module 201 is configured to acquire angiographic video of coronary angiography.
The image preprocessing module 202 is configured to perform video frame image extraction on the angiographic video to generate a corresponding first image sequence.
The stenosis blood vessel processing module 203 is configured to perform target detection and semantic segmentation processing on stenosis blood vessels in each first image in the first image sequence based on a preset image target detection and semantic segmentation model, so as to obtain one or more target detection frames for marking stenosis blood vessels and a stenosis blood vessel mask image in each target detection frame on each first image, where each target detection frame corresponds to a detection frame confidence.
The image optimization module 204 is configured to perform image optimization processing on the first image sequence according to the confidence level of the detection frame to obtain a specified number of preferred images.
The quantitative analysis module 205 is configured to perform coronary angiography quantitative analysis on the stenosis vessel mask image in each target detection frame on each preferred image to generate a corresponding vessel stenosis rate.
The device for performing quantitative analysis of coronary angiography based on angiography video provided by the embodiment of the invention can execute the method steps in the method embodiment, and has similar implementation principle and technical effects, and is not repeated herein.
It should be noted that, it should be understood that the division of the modules of the above apparatus is merely a division of a logic function, and may be fully or partially integrated into a physical entity or may be physically separated. The modules can be realized in the form of software which is called by the processing element, in the form of hardware, in the form of software which is called by the processing element, and in the form of hardware. For example, the acquisition module may be a processing element that is set up separately, may be implemented in a chip of the above apparatus, or may be stored in a memory of the above apparatus in the form of program code, and may be called by a processing element of the above apparatus and execute the functions of the above determination module. The implementation of the other modules is similar. In addition, all or part of the modules can be integrated together or can be independently implemented. The processing element described herein may be an integrated circuit having signal processing capabilities. In implementation, each step of the above method or each module above may be implemented by an integrated logic circuit of hardware in a processor element or an instruction in a software form.
For example, the modules above may be one or more integrated circuits configured to implement the methods above, such as one or more Application SPECIFIC INTEGRATED Circuits (ASICs), or one or more digital signal processors (DIGITAL SIGNAL processors, DSPs), or one or more field programmable gate arrays (Field Programmable GATE ARRAY, FPGA), or the like. For another example, when a module above is implemented in the form of processing element scheduler code, the processing element may be a general purpose processor, such as a central processing unit (Central Processing Unit, CPU) or other processor that may invoke the program code. For another example, the modules may be integrated together and implemented in the form of a System-on-a-chip (SOC).
In the above embodiments, it may be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, produces, in whole or in part, the processes or functions described in accordance with embodiments of the present invention. The computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, from one website, computer, server, or data center via a wired (e.g., coaxial cable, fiber optic, digital subscriber line (Digital Subscriber Line, DSL)) or wireless (e.g., infrared, wireless, bluetooth, microwave, etc.) means. The computer readable storage media may be any available media that can be accessed by a computer or a data storage device such as a server, data center, or the like that contains an integration of one or more available media. The usable medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a magnetic tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a Solid State Disk (SSD)), or the like.
Fig. 3 is a schematic structural diagram of an electronic device according to a third embodiment of the present invention. The electronic device may be the aforementioned terminal device or server, or may be a terminal device or server connected to the aforementioned terminal device or server for implementing the method of the embodiment of the present invention. As shown in fig. 3, the electronic device may include a processor 301 (e.g., CPU), a memory 302, and a transceiver 303, the transceiver 303 being coupled to the processor 301, the processor 301 controlling the transceiving actions of the transceiver 303. The memory 302 may store various instructions for performing various processing functions and implementing the methods and processes provided in the above-described embodiments of the present invention. Preferably, the electronic device according to the embodiment of the present invention further includes a power supply 304, a system bus 305, and a communication port 306. The system bus 305 is used to implement communication connections between the elements. The communication port 306 is used for connection communication between the electronic device and other peripheral devices.
The system bus referred to in fig. 3 may be a peripheral component interconnect standard (PERIPHERAL COMPONENT INTERCONNECT, PCI) bus, or an extended industry standard architecture (Extended Industry Standard Architecture, EISA) bus, or the like. The system bus may be classified into an address bus, a data bus, a control bus, and the like. For ease of illustration, the figures are shown with only one bold line, but not with only one bus or one type of bus. The communication interface is used to enable communication between the database access apparatus and other devices (e.g., clients, read-write libraries, and read-only libraries). The Memory may include random access Memory (Random Access Memory, RAM) and may also include Non-Volatile Memory (Non-Volatile Memory), such as at least one disk Memory.
The processor may be a general-purpose processor including a Central Processing Unit (CPU), a network processor (Network Processor, NP), etc., or may be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, a discrete gate or transistor logic device, or a discrete hardware component.
It should be noted that the embodiments of the present invention also provide a computer readable storage medium having instructions stored therein, which when executed on a computer, cause the computer to perform the methods and processes provided in the above embodiments.
The embodiment of the invention also provides a chip for running the instructions, which is used for executing the method and the processing procedure provided in the embodiment.
The embodiment of the invention provides a method, a device, electronic equipment and a computer-readable storage medium for quantitative analysis of coronary angiography based on angiography video, which are used for carrying out video interception and video frame image extraction processing on the angiography video of coronary vessels, carrying out target detection and semantic segmentation processing on narrow-section vessels on an extracted image sequence based on an image target detection and semantic segmentation model, optimizing an extracted image based on the confidence of target identification, and carrying out quantitative analysis of coronary angiography on each narrow-section vessel on the optimized image to generate corresponding vascular stenosis rate. The invention gets rid of the excessive dependence on manual experience in the traditional method and improves the image extraction accuracy and the stenosis rate calculation accuracy.
Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative elements and steps are described above generally in terms of function in order to clearly illustrate the interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied in hardware, in a software module executed by a processor, or in a combination of the two. The software modules may be disposed in Random Access Memory (RAM), memory, read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
The foregoing description of the embodiments has been provided for the purpose of illustrating the general principles of the invention, and is not meant to limit the scope of the invention, but to limit the invention to the particular embodiments, and any modifications, equivalents, improvements, etc. that fall within the spirit and principles of the invention are intended to be included within the scope of the invention.
Claims (7)
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210018415.6A CN114418977B (en) | 2022-01-07 | 2022-01-07 | Method and device for quantitative analysis of coronary angiography based on angiography video |
| PCT/CN2022/097245 WO2023130663A1 (en) | 2022-01-07 | 2022-06-07 | Coronary angiogram quantitative analysis method and apparatus based on angiogram video |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210018415.6A CN114418977B (en) | 2022-01-07 | 2022-01-07 | Method and device for quantitative analysis of coronary angiography based on angiography video |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114418977A CN114418977A (en) | 2022-04-29 |
| CN114418977B true CN114418977B (en) | 2025-05-27 |
Family
ID=81271742
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210018415.6A Active CN114418977B (en) | 2022-01-07 | 2022-01-07 | Method and device for quantitative analysis of coronary angiography based on angiography video |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN114418977B (en) |
| WO (1) | WO2023130663A1 (en) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114418977B (en) * | 2022-01-07 | 2025-05-27 | 乐普（北京）医疗器械股份有限公司 | Method and device for quantitative analysis of coronary angiography based on angiography video |
| CN114708259B (en) * | 2022-05-24 | 2022-08-16 | 深圳市铱硙医疗科技有限公司 | CTA (computed tomography angiography) -based head and neck vascular stenosis detection method, device, equipment and medium |
| CN115880227A (en) * | 2022-11-14 | 2023-03-31 | 北京理工大学 | Automatic detection method and system for coronary artery stenosis |
| CN116612102A (en) * | 2023-05-31 | 2023-08-18 | 上海博动医疗科技股份有限公司 | Blood vessel image processing system, device and storage medium |
| CN116958100A (en) * | 2023-07-27 | 2023-10-27 | 上海博动医疗科技股份有限公司 | A method, device, equipment and medium for extracting DSA sequence key frames |
| CN116703948B (en) * | 2023-08-03 | 2023-11-14 | 杭州脉流科技有限公司 | Intracranial vessel tree segmentation method and device based on deep neural network |
| CN116681716B (en) * | 2023-08-04 | 2023-10-10 | 杭州脉流科技有限公司 | Method, device, equipment and storage medium for dividing intracranial vascular region of interest |
| CN118000772B (en) * | 2023-12-05 | 2025-02-18 | 北京科莱博特医疗科技有限公司 | A blood flow fraction calculation method based on X-ray angiography image sequence |
| CN119784766B (en) * | 2025-03-13 | 2025-06-17 | 华中科技大学同济医学院附属协和医院 | Coronary artery vascular hemodynamic quantification method and system based on IVUS image |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109222980A (en) * | 2018-06-19 | 2019-01-18 | 北京红云智胜科技有限公司 | Method of the measurement coronarogram based on deep learning as blood vessel diameter |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9943277B2 (en) * | 2014-04-02 | 2018-04-17 | International Business Machines Corporation | Detecting coronary stenosis through spatio-temporal tracking |
| US11369278B2 (en) * | 2017-11-28 | 2022-06-28 | Beijing Keya Medical Technology Co., Ltd. | Methods and systems for determining lumen volume and coronary blood flow |
| CN111667456B (en) * | 2020-04-28 | 2023-11-14 | 北京理工大学 | A method and device for detecting vascular stenosis in coronary X-ray sequence angiography |
| CN111657883B (en) * | 2020-06-03 | 2021-05-04 | 北京理工大学 | Automatic calculation method and system of coronary SYNTAX score based on sequential angiography |
| CN111754506B (en) * | 2020-07-01 | 2024-02-06 | 杭州脉流科技有限公司 | Coronary artery stenosis rate calculation method, device and system based on intra-cavity image and computer storage medium |
| CN113628193B (en) * | 2021-08-12 | 2022-07-15 | 推想医疗科技股份有限公司 | Method, device and system for determining blood vessel stenosis rate and storage medium |
| CN113723419B (en) * | 2021-09-03 | 2023-11-21 | 乐普（北京）医疗器械股份有限公司 | Method and device for segmenting contrast image based on semantic segmentation network |
| CN113723418B (en) * | 2021-09-03 | 2023-08-04 | 乐普（北京）医疗器械股份有限公司 | Method and device for optimizing contrast image |
| CN114418977B (en) * | 2022-01-07 | 2025-05-27 | 乐普（北京）医疗器械股份有限公司 | Method and device for quantitative analysis of coronary angiography based on angiography video |
-
2022
- 2022-01-07 CN CN202210018415.6A patent/CN114418977B/en active Active
- 2022-06-07 WO PCT/CN2022/097245 patent/WO2023130663A1/en not_active Ceased
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109222980A (en) * | 2018-06-19 | 2019-01-18 | 北京红云智胜科技有限公司 | Method of the measurement coronarogram based on deep learning as blood vessel diameter |
Non-Patent Citations (1)
| Title |
|---|
| "一种基于Mask RCNN的融合几何特征的冠状动脉分割方法";邵凯 等;《太原理工大学学报》;20210131;第83-90页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2023130663A1 (en) | 2023-07-13 |
| CN114418977A (en) | 2022-04-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114418977B (en) | Method and device for quantitative analysis of coronary angiography based on angiography video | |
| CN112419271B (en) | Image segmentation method, device and computer readable storage medium | |
| Balakrishna et al. | Automatic detection of lumen and media in the IVUS images using U-Net with VGG16 Encoder | |
| CN114419499B (en) | An ejection fraction data processing method and device based on cardiac ultrasound video | |
| WO2023137948A1 (en) | Processing method and apparatus for analyzing fractional flow reserve on the basis of angiographic image | |
| CN114418976B (en) | Angiographic image analysis method and device | |
| CN114419500B (en) | Method and device for screening diastole and systole images based on heart ultrasonic video | |
| CN108335284A (en) | A kind of coronary artery vessel centerline matching process and system | |
| CN111915557B (en) | A deep learning method and device for detecting atrial septal defect | |
| CN107122735B (en) | Multi-target tracking method based on deep learning and conditional random field | |
| WO2023130661A1 (en) | Method and apparatus for processing two-dimensional spectral doppler echocardiographic image | |
| CN113538277B (en) | A method and device for reducing noise of tomographic images based on neural network | |
| CN110379491B (en) | Method, device, equipment and storage medium for identifying brain glioma | |
| CN113139568A (en) | Class prediction model modeling method and device based on active learning | |
| Mei et al. | Cross-stained segmentation from renal biopsy images using multi-level adversarial learning | |
| CN117132974A (en) | Object detection methods, devices and electronic equipment based on artificial intelligence | |
| CN114010227B (en) | Right ventricle characteristic information identification method and device | |
| CN113723419B (en) | Method and device for segmenting contrast image based on semantic segmentation network | |
| CN110717916B (en) | Pulmonary embolism detection system based on convolutional neural network | |
| CN114998582A (en) | Coronary blood vessel segmentation method, device and storage medium | |
| CN111292298A (en) | Breast cancer pathological typing determination method, device and storage medium | |
| CN112308844A (en) | Blood vessel lumen extraction method and device, electronic equipment and storage medium | |
| CN114511526A (en) | Aneurysm segmentation method and device based on deep learning | |
| CN118467763B (en) | Processing method and device for constructing DWI image medical atlas based on image registration | |
| CN118172335B (en) | Processing method and device for evaluating and applying CATCH sequence scheme |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |