CN111862038B - Plaque detection method, plaque detection device, plaque detection equipment and plaque detection medium - Google Patents
Plaque detection method, plaque detection device, plaque detection equipment and plaque detection medium Download PDFInfo
- Publication number
- CN111862038B CN111862038B CN202010693473.XA CN202010693473A CN111862038B CN 111862038 B CN111862038 B CN 111862038B CN 202010693473 A CN202010693473 A CN 202010693473A CN 111862038 B CN111862038 B CN 111862038B
- Authority
- CN
- China
- Prior art keywords
- plaque
- coronary artery
- detected
- final
- image
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 236
- 210000004351 coronary vessel Anatomy 0.000 claims abstract description 279
- 238000000034 method Methods 0.000 claims abstract description 14
- 208000031481 Pathologic Constriction Diseases 0.000 claims description 14
- 208000037804 stenosis Diseases 0.000 claims description 14
- 230000036262 stenosis Effects 0.000 claims description 14
- 238000003709 image segmentation Methods 0.000 claims description 13
- 238000000605 extraction Methods 0.000 claims description 6
- 206010057469 Vascular stenosis Diseases 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000010606 normalization Methods 0.000 claims description 4
- 238000004590 computer program Methods 0.000 claims description 2
- 238000005259 measurement Methods 0.000 claims 2
- 230000000694 effects Effects 0.000 abstract description 8
- 238000005516 engineering process Methods 0.000 abstract description 7
- 230000011218 segmentation Effects 0.000 description 20
- 210000004204 blood vessel Anatomy 0.000 description 18
- 230000006870 function Effects 0.000 description 10
- 238000012549 training Methods 0.000 description 9
- 238000010586 diagram Methods 0.000 description 4
- 208000029078 coronary artery disease Diseases 0.000 description 3
- 206010003211 Arteriosclerosis coronary artery Diseases 0.000 description 2
- 101000666730 Homo sapiens T-complex protein 1 subunit alpha Proteins 0.000 description 2
- 102100038410 T-complex protein 1 subunit alpha Human genes 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 208000026758 coronary atherosclerosis Diseases 0.000 description 2
- 238000013136 deep learning model Methods 0.000 description 2
- 230000003902 lesion Effects 0.000 description 2
- 238000000926 separation method Methods 0.000 description 2
- 206010021143 Hypoxia Diseases 0.000 description 1
- 206010028851 Necrosis Diseases 0.000 description 1
- 238000002583 angiography Methods 0.000 description 1
- 210000001367 artery Anatomy 0.000 description 1
- 230000003143 atherosclerotic effect Effects 0.000 description 1
- 239000008280 blood Substances 0.000 description 1
- 210000004369 blood Anatomy 0.000 description 1
- 238000010968 computed tomography angiography Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 230000007954 hypoxia Effects 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 208000031225 myocardial ischemia Diseases 0.000 description 1
- 230000017074 necrotic cell death Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
- 230000002861 ventricular Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/12—Edge-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/03—Recognition of patterns in medical or anatomical images
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Radiology & Medical Imaging (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Medical Informatics (AREA)
- Geometry (AREA)
- Probability & Statistics with Applications (AREA)
- Quality & Reliability (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The embodiment of the invention discloses a plaque detection method, a plaque detection device, plaque detection equipment and a plaque detection medium. The method comprises the following steps: acquiring a coronary artery image sequence to be detected; generating at least one coronary artery image combination to be detected according to the coronary artery image sequence to be detected, wherein the coronary artery image combination to be detected comprises at least two coronary artery images to be detected; inputting the coronary artery image combination to be detected into a trained plaque detection model to obtain an intermediate plaque detection result of each coronary artery image to be detected; and determining a final plaque detection result based on the intermediate plaque detection result of each coronary artery image to be detected. The technical scheme of the embodiment of the invention solves the problems of high detection difficulty and inaccurate detection of the existing coronary plaque detection technology, and achieves the effects of reducing the detection difficulty of the coronary plaque and improving the plaque detection efficiency and accuracy.
Description
Technical Field
The embodiment of the invention relates to a medical image technology, in particular to a plaque detection method, a plaque detection device, plaque detection equipment and a plaque detection medium.
Background
Coronary arteries are arteries responsible for supplying blood to the heart, and when they undergo atherosclerotic lesions, they cause stenosis or blockage of the lumen of the blood vessel, which in turn causes myocardial ischemia, hypoxia or necrosis, also commonly known as "coronary heart disease".
Coronary artery computed tomography angiography (Coronary Computed Tomographic Angiography, CCTA) is a means of conventional coronary examination by which the location, nature, and extent of stenosis of a patient's coronary atherosclerosis are determined. Coronary atherosclerosis is classified into calcified plaque, non-calcified plaque, and mixed plaque having both calcified and non-calcified properties.
The method for judging the nature of the coronary plaque is commonly used at present: firstly, carrying out coronary artery extraction and coronary artery separation based on CCTA data, and then reconstructing the separated coronary arteries one by utilizing a reconstruction technology and judging the stenosis position and plaque property. The effect of reconstructing the coronary artery depends on the effect of extracting and separating the coronary artery, and the difficulty of extracting and separating the tiny branch coronary artery is high, and the method is very time-consuming, especially when the stenosis rate of a coronary artery at a certain place reaches 100%, the coronary artery is completely blocked, and the result shown on the image is that one coronary artery is truncated into two parts, at the moment, the second coronary artery may not be extracted, and the subsequent coronary artery reconstruction also fails. Therefore, the existing method for judging the nature of the coronary plaque has the problems of great difficulty and low efficiency.
Disclosure of Invention
The embodiment of the invention provides a plaque detection method, device, equipment and medium, which are used for realizing the effects of reducing the difficulty of coronary plaque detection and improving the plaque detection efficiency.
In a first aspect, an embodiment of the present invention provides a plaque detection method, including:
Acquiring a coronary artery image sequence to be detected;
Generating at least one coronary artery image combination to be detected according to the coronary artery image sequence to be detected, wherein the coronary artery image combination to be detected comprises at least two coronary artery images to be detected;
inputting the coronary artery image combination to be detected into a trained plaque detection model to obtain an intermediate plaque detection result of each coronary artery image to be detected;
And determining a final plaque detection result based on the intermediate plaque detection result of each coronary artery image to be detected.
In a second aspect, an embodiment of the present invention further provides a plaque detection apparatus, including:
The image sequence acquisition module is used for acquiring a coronary artery image sequence to be detected;
the image combination generating module is used for generating at least one coronary artery image combination to be detected according to the coronary artery image sequence to be detected, and the coronary artery image combination to be detected comprises at least two coronary artery images to be detected;
The intermediate plaque detection result acquisition module is used for inputting the coronary artery image combination to be detected into a trained plaque detection model to obtain an intermediate plaque detection result of each coronary artery image to be detected;
And the final plaque detection result determining module is used for determining a final plaque detection result based on the intermediate plaque detection result of each coronary artery image to be detected.
In a third aspect, an embodiment of the present invention further provides an apparatus, where the apparatus includes:
one or more processors;
A storage means for storing one or more programs;
The one or more programs, when executed by the one or more processors, cause the one or more processors to implement the plaque detection method as provided by any embodiment of the invention.
In a fourth aspect, embodiments of the present invention further provide a computer readable storage medium having a computer program stored thereon, wherein the program when executed by a processor implements the plaque detection method as provided by any of the embodiments of the present invention.
The embodiment of the invention obtains the coronary artery image sequence to be detected; generating at least one coronary artery image combination to be detected according to the coronary artery image sequence to be detected, wherein the coronary artery image combination to be detected comprises at least two coronary artery images to be detected; inputting the coronary artery image combination to be detected into a trained plaque detection model to obtain an intermediate plaque detection result aiming at the coronary artery image combination to be detected; and determining a final plaque detection result based on the intermediate plaque detection result, so that the problems of high detection difficulty and inaccurate detection in the existing coronary plaque detection technology are solved, and the effects of reducing the detection difficulty of the coronary plaque and improving the plaque detection efficiency and accuracy are realized.
Drawings
Fig. 1 is a flowchart of a plaque detection method according to a first embodiment of the present invention;
FIG. 2 is a schematic diagram of plaque detection and vessel segmentation in accordance with a first embodiment of the present invention;
fig. 3 is a flowchart of a plaque detection method according to a second embodiment of the present invention;
fig. 4 is a block diagram of a plaque detection apparatus in a third embodiment of the present invention;
Fig. 5 is a schematic structural view of an apparatus according to a fourth embodiment of the present invention.
Detailed Description
The invention is described in further detail below with reference to the drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting thereof. It should be further noted that, for convenience of description, only some, but not all of the structures related to the present invention are shown in the drawings.
Example 1
Fig. 1 is a flowchart of a plaque detection method according to an embodiment of the present invention, where the embodiment is applicable to plaque detection, and the method may be performed by a plaque detection apparatus, and specifically includes the following steps:
S110, acquiring a coronary artery image sequence to be detected.
The method comprises the steps of obtaining an image sequence obtained after a medical image device scans coronary arteries, carrying out windowing pretreatment on the obtained coronary artery image sequence, compressing pixel values of the image within a range of 0-255, enabling only information required by the coronary arteries and the like to be displayed in the image sequence, avoiding interference of other useless information on plaque detection of the coronary arteries, improving the plaque detection accuracy, compressing the pixel value range of the image within a preset range, and avoiding inaccurate plaque detection model detection caused by overlarge pixel value change range, so that the plaque detection accuracy is reduced.
S120, generating at least one coronary artery image combination to be detected according to the coronary artery image sequence to be detected, wherein the coronary artery image combination to be detected comprises at least two coronary artery images to be detected.
And combining the adjacent preset number of images in the coronary artery image sequence to be detected into at least one coronary artery image combination to be detected, wherein the preset number is at least two. At least one image combination is input into the plaque detection model. Optionally, generating at least one coronary artery image combination to be detected according to the coronary artery image sequence to be detected includes: starting from a first coronary artery image to be detected of the coronary artery image sequence to be detected, selecting adjacent preset number of coronary artery images to be detected as a first coronary artery image combination to be detected; starting from a second coronary artery image to be detected of the coronary artery image sequence to be detected, selecting adjacent preset number of coronary artery images to be detected as a second coronary artery image combination to be detected; repeating the operation until each coronary artery image to be detected of the coronary artery image sequence to be detected is traversed. The acquired coronary artery image sequence to be detected comprises 20 images, and the 1 st to 3 rd coronary artery images to be detected are sequentially selected from the first coronary artery image to be detected for the first time to obtain a first coronary artery image combination to be detected; starting from the second coronary artery image to be detected, sequentially selecting the 2 nd to 4 th coronary artery images to be detected to obtain a second coronary artery image combination to be detected; repeating the above operation, selecting 3 adjacent coronary artery images to be detected each time to obtain a coronary artery image combination to be detected, until each coronary artery image to be detected in the coronary artery image sequence to be detected is traversed, so as to obtain all image combinations of the coronary artery image sequence to be detected, wherein the image combinations are generated as shown in fig. 2 (a). And combining the images in the coronary artery image sequence to be detected to generate images, and inputting the images into the plaque detection model, so that the detection model can obtain more image information, the plaque detection in the images can be facilitated by the detection model, and the plaque detection is more accurate.
S130, inputting the coronary artery image combination to be detected into a trained plaque detection model, and obtaining an intermediate plaque detection result of each coronary artery image to be detected.
And inputting the coronary artery image combination to be detected into a plaque detection model to obtain plaque detection results of the coronary artery images to be detected in the middle sequence in the coronary artery image combination to be detected. In order to enable the plaque detection model to obtain more coronary artery image information, the coronary artery image to be detected is generated into a coronary artery image combination to be detected, and the coronary artery image combination is input into the plaque detection model, so that the obtained plaque detection result is more accurate.
The plaque detection model is trained in advance and is used for detecting plaque properties and plaque positions of the coronary artery images to be detected; because a large number of focuses are concentrated on a plurality of thick branches such as anterior descending branches, convolution branches and the like, and the number of focus samples on the tiny branches is relatively small, when a plaque detection model is trained, the tiny branch focuses are required to be subjected to data enhancement, including means of translation, rotation and the like, the number proportion of the tiny branch focus samples in the training samples is increased, and therefore the detection accuracy of the plaque detection model on the focuses on the tiny branches is improved.
Plaque properties include: calcified plaque, non-calcified plaque, and mixed plaque. Because the non-calcified plaque sample is less, the difficulty of learning the plaque detection model is higher, and the duty ratio of the non-calcified plaque in the training sample is required to be improved. The plaque detection model with short training time is used, and because the training time is short, the plaque detection model can generate a large amount of non-calcified plaque false positive data, and the non-calcified plaque false positive data is used as input data to be added into a training data set of the plaque detection model, so that the plaque detection model is optimized, and the detection accuracy of the plaque detection model on the non-calcified plaque is improved.
Since the morphology of different types of blood vessels is different, the plaque morphology in the blood vessels of different morphologies is also different. Because the morphology of the distal end and the left ventricular posterior branch in the anterior descending branch of individual blood vessels is very similar in some cases, the cavity convolution is used in the plaque detection model design, the receptive field of the model is increased on the premise of not changing the depth of the model, the morphology of the blood vessels is easier to distinguish, and the plaque detection model can detect the plaque of different types of blood vessels more accurately.
Optionally, inputting the coronary artery image combination to be detected into a trained plaque detection model, including: normalizing the spatial coordinate values of the intermediate sequence coronary artery images to be detected in the coronary artery image combination to be detected; and inputting the spatial coordinate values after the coronary artery image combination and normalization to a trained plaque detection model. Because of individual differences of people, different individual coronary artery images shot by using image equipment have different resolutions, so that the difference of spatial coordinate values of the coronary artery images of different individuals is larger, and the plaque detection model is not beneficial to positioning the plaque in the coronary artery images, therefore, the spatial coordinate values of the images need to be normalized, the numerical range of the spatial coordinate values is between 0 and 1, and the accuracy of the plaque detection model on plaque position detection is improved. The plaque detection model outputs a detection result which is the plaque detection result of the intermediate-order coronary artery images in the input coronary artery image combination to be detected, so that only the space coordinate values of the intermediate-order coronary artery images in the coronary artery image combination to be detected are calculated, and the space coordinate values of the intermediate-order coronary artery images in the coronary artery image combination to be detected and each image combination are input into the plaque detection model to obtain the intermediate plaque detection result of the intermediate-order coronary artery images in each coronary artery image combination to be detected. Illustratively, as shown in fig. 2 (b), the intermediate plaque detection results are obtained, where the location of the box is the plaque location, rca_d indicates that the lesion location is the distal end of the right coronary artery, ncP indicates that the plaque is non-calcified.
And S140, determining a final plaque detection result based on the intermediate plaque detection result of each coronary artery image to be detected.
The intermediate plaque detection result output by the plaque detection model is the detection result of the intermediate sequence coronary artery images to be detected in each coronary artery image combination to be detected, and is the plaque detection result of the two-dimensional image. And determining a final plaque detection result through the detection results of the intermediate sequence coronary artery images to be detected in each coronary artery image combination to be detected.
Optionally, determining the final plaque detection result based on the intermediate plaque detection result for each coronary artery image to be detected includes: determining the intersection ratio between plaque areas in every two coronary artery images to be detected; clustering the first plaque positions in the coronary artery image to be detected, wherein the intersection ratio exceeds a preset threshold value, so as to obtain the final plaque positions in the coronary artery image to be detected; wherein the intermediate plaque detection result includes: and a first plaque position, wherein the final plaque detection result comprises a final plaque position. The first plaque position is a two-dimensional plaque position, and the final plaque position is a three-dimensional plaque position. And calculating an intersection and a union of the plaque position of each coronary artery image to be detected in the intermediate plaque detection results, namely the image area corresponding to the first plaque position, and the image areas corresponding to the plaque positions of other coronary artery images to be detected, and calculating the ratio of the intersection to the union. The plaque detection results of intermediate sequential images in each combination are obtained through a plaque detection model, wherein plaque positions corresponding to the combination A, the combination B and the combination C are respectively a position 1, a position 2 and a position 3, intersection and union calculation are carried out on image areas corresponding to the position 1 and the position 2 and the position 3 respectively to obtain an intersection 12, an intersection 13, a union 12 and a union 13, the intersection corresponds to the union one by one, the ratio of the intersection 12 to the union 12 is calculated, the ratio of the intersection 13 to the union 13 is calculated, the position 1 and the position 2 are clustered if the ratio of the intersection 12 to the union 12 exceeds a preset threshold, and the position 1 and the position 3 are clustered if the ratio of the intersection 13 to the union 13 exceeds a preset threshold, and the position 1, the position 2 and the position 3 are clustered because the position 1 and the position 2 are clustered; if the ratio of intersection 13 to union 13 does not exceed the preset threshold, then location 3 cannot be clustered with location 1 and location 2. As shown in fig. 2 (c), the three-dimensional plaque positions obtained by clustering the two-dimensional plaque positions in the intermediate plaque detection result are final plaque positions.
Optionally, determining the final plaque detection result based on the intermediate plaque detection result for each coronary artery image to be detected further includes: counting the plaque quantity of each plaque property in the coronary artery image to be detected, wherein the intersection ratio exceeds a preset threshold; determining final properties of the plaque based on the number of plaques of each type of plaque property; wherein, the intermediate plaque detection result includes: each type of property of the plaque, the final plaque detection results include the final property of the plaque. Classifying the plaque properties corresponding to each plaque position clustered together, counting the number of the plaque properties of each type, and determining the final properties of the clustered plaques according to the number of the plaque properties of each type.
Optionally, determining the final property of the plaque for the plaque number of each type of plaque property includes: sorting the plaque numbers of different properties in a descending order; if the ratio of the minimum number of the two types of plaques to the total number of the plaques with different properties is larger than a preset ratio, determining that the final property of the plaques is a mixed plaque; and if the ratio of the number of the plaques with the smallest number and the total number of the plaques with different properties is smaller than the preset ratio, determining the plaque property with the largest number as the final property of the plaque. Illustratively, the predetermined ratio is 0.4, 6 plaque sites are clustered together, 4 calcified plaques, 2 mixed plaques and 0 non-calcified plaques, the two smallest categories are mixed plaques and non-calcified plaques, the sum of the numbers is 2, 0.33 and 0.33 is less than 0.4 of the total number, so that the final plaque property of the three-dimensional plaque sites clustered together is calcified plaques. If the 6 clustered plaque sites include 3 calcified plaques, 2 mixed plaques and 1 non-calcified plaque, the two categories with the smallest number are mixed plaques and non-calcified plaques, the sum of the two categories is 3, and 0.50,0.50 of the total number is greater than 0.4, and the final plaque property of the clustered three-dimensional plaque sites is the mixed plaque.
The plaque detection model can directly detect the plaque of the original coronary artery image, the position and the property of the plaque are not required to be judged through the reconstructed blood vessel after the blood vessel separation and reconstruction are carried out on the coronary artery image, the time cost and the calculation cost required by three-dimensional reconstruction are reduced, meanwhile, the interference of reconstruction noise on the detection result can be effectively avoided, and the plaque detection efficiency and accuracy are improved.
According to the technical scheme, a coronary artery image sequence to be detected is obtained; generating at least one coronary artery image combination to be detected according to the coronary artery image sequence to be detected, wherein the coronary artery image combination to be detected comprises at least two coronary artery images to be detected; inputting the coronary artery image combination to be detected into a trained plaque detection model to obtain an intermediate plaque detection result aiming at the coronary artery image combination to be detected; and determining a final plaque detection result based on the intermediate plaque detection result, so that the problems of high detection difficulty and inaccurate detection in the existing coronary plaque detection technology are solved, and the effects of reducing the detection difficulty of the coronary plaque and improving the plaque detection efficiency and accuracy are realized.
Example two
Fig. 3 is a flowchart of a plaque detection method according to a second embodiment of the present invention, where the plaque detection method is further optimized for the previous embodiment, and the plaque detection method further includes: image segmentation is carried out on the coronary artery image to be detected according to the final plaque position included in the final plaque detection result, so as to extract the coronary artery blood vessel where the plaque is located; the vascular stenosis rate is calculated from the extracted coronary vessels. After plaque detection is carried out on the coronary artery image, the coronary artery image with plaque is subjected to image segmentation to extract coronary artery blood vessels where the plaque is located, and the blood vessel stenosis rate is calculated, so that the image segmentation of the coronary artery image without plaque is avoided, and the time and difficulty of blood vessel extraction are effectively reduced.
As shown in fig. 3, the method comprises the following steps:
S210, acquiring a coronary artery image sequence to be detected.
S220, generating at least one coronary artery image combination to be detected according to the coronary artery image sequence to be detected, wherein the coronary artery image combination to be detected comprises at least two coronary artery images to be detected.
S230, inputting the coronary artery image combination to be detected into a trained plaque detection model, and obtaining an intermediate plaque detection result of each coronary artery image to be detected.
S240, determining a final plaque detection result based on the intermediate plaque detection result of each coronary artery image to be detected.
S250, image segmentation is carried out on the coronary artery image to be detected according to the final plaque position included in the final plaque detection result, so as to extract the coronary artery blood vessel where the plaque is located.
In order to accelerate the convergence rate of the coronary artery segmentation model, the coronary artery images corresponding to the clustered plaque positions are subjected to data enhancement operations such as isotropic transformation based on spatial resolution, rotation, scaling, contrast change and the like, and are input into the trained coronary artery segmentation model for image segmentation, so that coronary artery blood vessels where the plaque is located are extracted. Alternatively, training the coronary segmentation model may be: and inputting the sample image into a coronary artery segmentation model to be trained, and obtaining a prediction result output by the detection model, wherein the prediction result comprises a predicted blood vessel segmentation result. The loss function of the coronary artery segmentation model to be trained comprises a cross entropy loss function and a loss function based on a contour, the contour precision determines the accuracy of the extracted blood vessel diameter, and therefore the segmentation module focuses more on the contour precision of the segmentation result, and besides the traditional cross entropy loss function, the loss function based on the contour is used in the coronary artery segmentation model, so that the accuracy of the coronary artery segmentation model on blood vessel segmentation is improved. The cross entropy loss function calculates a predicted vessel segmentation result pixel by pixel that is different from a standard vessel segmentation result. The predicted vessel segmentation result calculated based on the loss function of the profile is different from the vessel profile of the standard vessel segmentation result. And reversely inputting the loss function into a detection model to be trained, and adjusting network parameters in the detection model based on a gradient descent method. And iteratively executing the training method until the training of the preset times is completed or the detection precision of the detection model reaches the preset precision, and determining that the training of the coronary artery segmentation model is completed.
And S260, calculating the vascular stenosis rate according to the extracted coronary artery blood vessel.
Determining the midline of the coronary artery blood vessel by midline extraction technology, measuring the distance between the edge of the blood vessel and the midline, wherein the distance is 2 times of the caliber of the coronary artery blood vessel; measuring the real-time pipe diameter of a coronary artery blood vessel, presetting a standard pipe diameter, and calculating the pipe diameter ratio between the real-time pipe diameter and the preset standard pipe diameter; the stenosis rate of the coronary artery blood vessel is determined by the pipe diameter ratio.
On the basis of the above embodiment, the technical solution of the embodiment of the present invention further includes: the plaque detection model and the coronary artery segmentation model are realized through a deep learning model, a two-dimensional coronary artery image is input into the deep learning model, plaque positions and plaque properties can be output, segmentation of blood vessels corresponding to the plaque positions is completed at the same time, the operations of clustering the plaque positions to determine final plaque positions and counting the quantity of each type of plaque properties to determine final plaque properties are not needed, and the coronary artery plaque detection and blood vessel segmentation efficiency is improved.
According to the technical scheme, a coronary artery image sequence to be detected is obtained; generating at least one coronary artery image combination to be detected according to the coronary artery image sequence to be detected, wherein the coronary artery image combination to be detected comprises at least two coronary artery images to be detected; inputting the coronary artery image combination to be detected into a trained plaque detection model to obtain an intermediate plaque detection result aiming at the coronary artery image combination to be detected; determining a final plaque detection result based on the intermediate plaque detection result, and performing image segmentation on the coronary artery image to be detected according to the final plaque position included in the final plaque detection result so as to extract the coronary artery blood vessel where the plaque is located; the vascular stenosis rate is calculated from the extracted coronary vessels. After plaque detection is carried out on the coronary artery image, the coronary artery image with plaque is subjected to image segmentation to extract coronary artery blood vessels where the plaque is located, and the blood vessel stenosis rate is calculated, so that the image segmentation of the coronary artery image without plaque is avoided, and the time and difficulty of blood vessel extraction are effectively reduced. The problems of high detection difficulty and inaccurate detection of the existing coronary plaque detection technology are solved, and the effects of reducing the detection difficulty of the coronary plaque and improving the plaque detection efficiency and accuracy are achieved.
Example III
Fig. 4 is a block diagram of a plaque detection apparatus according to a third embodiment of the present invention, where the plaque detection apparatus includes: an image sequence acquisition module 310, an image combination generation module 320, an intermediate plaque detection result acquisition module 330, and a final plaque detection result determination module 340.
The image sequence acquiring module 310 is configured to acquire a coronary artery image sequence to be detected;
an image combination generating module 320, configured to generate at least one coronary artery image combination to be detected according to the coronary artery image sequence to be detected, where the coronary artery image combination to be detected includes at least two coronary artery images to be detected;
the intermediate plaque detection result obtaining module 330 is configured to input the combination of the coronary artery images to be detected to a trained plaque detection model, and obtain an intermediate plaque detection result of each coronary artery image to be detected;
A final plaque detection result determining module 340, configured to determine a final plaque detection result based on the intermediate plaque detection result of each coronary artery image to be detected.
In the technical solution of the foregoing embodiment, the image combination generating module 320 includes:
An image combination selecting unit, configured to select, from a first coronary artery image to be detected in the coronary artery image sequence to be detected, a preset number of adjacent coronary artery images to be detected as a first coronary artery image combination to be detected; starting from a second coronary artery image to be detected of the coronary artery image sequence to be detected, selecting adjacent preset number of coronary artery images to be detected as a second coronary artery image combination to be detected; repeating the operation until each coronary artery image to be detected of the coronary artery image sequence to be detected is traversed.
In the technical solution of the foregoing embodiment, the final plaque detection result determining module 340 includes:
The area intersection ratio determining unit is used for determining the intersection ratio between plaque areas in each two coronary artery images to be detected;
The plaque position clustering unit is used for clustering the first plaque position in the coronary artery image to be detected, wherein the intersection ratio of the first plaque position and the second plaque position exceeds a preset threshold value, and the final plaque position in the coronary artery image to be detected is obtained; wherein the intermediate plaque detection result includes: and a first plaque position, wherein the final plaque detection result comprises a final plaque position.
In the technical solution of the foregoing embodiment, the final plaque detection result determining module 340 further includes:
The plaque quantity counting unit is used for counting the plaque quantity of each plaque property in the coronary artery image to be detected, wherein the plaque quantity exceeds a preset threshold;
A final plaque property determining unit configured to determine a final property of the plaque according to the plaque number of each type of plaque property; wherein the intermediate plaque detection result includes: each type of property of the plaque, the final plaque detection results include the final property of the plaque.
In the technical solution of the above embodiment, the final plaque property determining unit includes:
The plaque sorting subunit is used for sorting the plaque numbers with different properties in a descending order;
A final plaque property determining subunit, configured to determine that the final property of the plaque is a mixed plaque if a ratio of the number of the two types of plaques with the smallest number to the total number of plaques with different properties is greater than a preset ratio; and if the ratio of the number of the plaques with the smallest number and the total number of the plaques with different properties is smaller than the preset ratio, determining the plaque property with the largest number as the final property of the plaque.
In the solution of the foregoing embodiment, the intermediate plaque detection result obtaining module 330 includes:
The coordinate value normalization unit is used for normalizing the spatial coordinate values of the intermediate sequence coronary artery images to be detected in the coronary artery image combination to be detected;
And the data input unit is used for inputting the spatial coordinate values after the coronary artery image combination and normalization to the trained plaque detection model.
In the technical solution of the foregoing embodiment, the plaque detection apparatus further includes:
The image segmentation module is used for carrying out image segmentation on the coronary artery image to be detected according to the final plaque position included in the final plaque detection result so as to extract the coronary artery blood vessel where the plaque is located;
and the stenosis rate calculation module is used for calculating the vessel stenosis rate according to the extracted coronary vessels.
According to the technical scheme, a coronary artery image sequence to be detected is obtained; generating at least one coronary artery image combination to be detected according to the coronary artery image sequence to be detected, wherein the coronary artery image combination to be detected comprises at least two coronary artery images to be detected; inputting the coronary artery image combination to be detected into a trained plaque detection model to obtain an intermediate plaque detection result aiming at the coronary artery image combination to be detected; and determining a final plaque detection result based on the intermediate plaque detection result, so that the problems of high detection difficulty and inaccurate detection in the existing coronary plaque detection technology are solved, and the effects of reducing the detection difficulty of the coronary plaque and improving the plaque detection efficiency and accuracy are realized.
The plaque detection device provided by the embodiment of the invention can execute the plaque detection method provided by any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.
Example IV
Fig. 5 is a schematic structural diagram of an apparatus according to a fourth embodiment of the present invention, and as shown in fig. 5, the apparatus includes a processor 410, a memory 420, an input device 430 and an output device 440; the number of processors 410 in the device may be one or more, one processor 410 being taken as an example in fig. 5; the processor 410, memory 420, input means 430 and output means 440 in the device may be connected by a bus or other means, for example by a bus connection in fig. 5.
The memory 420 is used as a computer readable storage medium, and may be used to store software programs, computer executable programs, and modules, such as program instructions/modules corresponding to the plaque detection method in the embodiment of the present invention (for example, the image sequence acquisition module 310, the image combination generation module 320, the intermediate plaque detection result acquisition module 330, and the final plaque detection result determination module 340 in the plaque detection apparatus). The processor 410 executes various functional applications of the device and data processing, i.e., implements the plaque detection method described above, by running software programs, instructions, and modules stored in the memory 420.
Memory 420 may include primarily a program storage area and a data storage area, wherein the program storage area may store an operating system, at least one application program required for functionality; the storage data area may store data created according to the use of the terminal, etc. In addition, memory 420 may include high-speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid-state storage device. In some examples, memory 420 may further include memory located remotely from processor 410, which may be connected to the device via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The input means 430 may be used to receive entered numeric or character information and to generate key signal inputs related to user settings and function control of the device. The output 440 may include a display device such as a display screen.
Example five
A fifth embodiment of the present invention also provides a storage medium containing computer-executable instructions, which when executed by a computer processor, are for performing a plaque detection method, the method comprising:
Acquiring a coronary artery image sequence to be detected;
Generating at least one coronary artery image combination to be detected according to the coronary artery image sequence to be detected, wherein the coronary artery image combination to be detected comprises at least two coronary artery images to be detected;
inputting the coronary artery image combination to be detected into a trained plaque detection model to obtain an intermediate plaque detection result of each coronary artery image to be detected;
And determining a final plaque detection result based on the intermediate plaque detection result of each coronary artery image to be detected.
Of course, the storage medium containing the computer executable instructions provided in the embodiments of the present invention is not limited to the method operations described above, and may also perform the related operations in the plaque detection method provided in any embodiment of the present invention.
From the above description of embodiments, it will be clear to a person skilled in the art that the present invention may be implemented by means of software and necessary general purpose hardware, but of course also by means of hardware, although in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as a floppy disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a FLASH Memory (FLASH), a hard disk, or an optical disk of a computer, etc., and include several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method according to the embodiments of the present invention.
It should be noted that, in the above embodiment of the plaque detection apparatus, each unit and module included are only divided according to the functional logic, but not limited to the above 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.
Note that the above is only a preferred embodiment of the present invention and the technical principle applied. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, while the invention has been described in connection with the above embodiments, the invention is not limited to the embodiments, but may be embodied in many other equivalent forms without departing from the spirit or scope of the invention, which is set forth in the following claims.
Claims (6)
1. A plaque detection method, comprising:
Acquiring a coronary artery image sequence to be detected;
Generating at least one coronary artery image combination to be detected according to the coronary artery image sequence to be detected, wherein the coronary artery image combination to be detected comprises at least two coronary artery images to be detected;
Inputting the combination of the coronary artery images to be detected into a trained plaque detection model to obtain an intermediate plaque detection result of each coronary artery image to be detected, wherein the plaque detection model is trained in advance and is used for detecting plaque properties and plaque positions of the coronary artery images to be detected, and the plaque properties comprise: calcified plaque, non-calcified plaque, and mixed plaque;
Determining a final plaque detection result based on the intermediate plaque detection result of each coronary artery image to be detected, including:
Determining the intersection ratio between plaque areas in every two coronary artery images to be detected; clustering the first plaque positions in the coronary artery image to be detected, wherein the intersection ratio exceeds a preset threshold value, so as to obtain the final plaque positions in the coronary artery image to be detected; wherein the intermediate plaque detection result includes: the plaque detection method comprises the steps that a plaque detection result comprises a final plaque position, wherein the first plaque position is a two-dimensional plaque position, and the final plaque position is a three-dimensional plaque position;
Determining a final plaque detection result based on the intermediate plaque detection result of each coronary artery image to be detected, further comprising: counting the plaque quantity of each plaque property in the coronary artery image to be detected, wherein the intersection ratio exceeds a preset threshold; determining final properties of the plaque according to the plaque number of each type of plaque properties; wherein the intermediate plaque detection result includes: each type of property of the plaque, the final plaque detection results including final properties of the plaque;
image segmentation is carried out on the coronary artery image to be detected according to the final plaque position included in the final plaque detection result so as to extract the coronary artery blood vessel where the plaque is located; calculating a vascular stenosis rate from the extracted coronary vessels;
wherein said determining final properties of said plaque from said plaque number of each type of plaque properties comprises:
The plaque number of different properties is ordered in a descending order, wherein the ordering order of the plaque of different properties is as follows: calcified plaque, mixed plaque, non-calcified plaque; if the ratio of the number of the plaques with the two types of properties to the total number of the plaques is larger than a preset ratio, determining that the final property of the plaques is a mixed plaque; if the ratio of the number of the two types of the plaques to the total number of the plaques is smaller than a preset ratio, determining the final property of the plaques to be calcified plaques, wherein the two types of the plaques are mixed plaques and non-calcified plaques;
wherein the calculating the vascular stenosis rate from the extracted coronary vessels comprises:
determining the midline of the coronary artery blood vessel by a midline extraction technology, and measuring the distance from the edge of the coronary artery blood vessel to the midline, wherein the caliber of the coronary artery blood vessel is twice the distance from the edge of the coronary artery blood vessel to the midline;
measuring the real-time pipe diameter of the coronary artery blood vessel in the measurement mode, presetting a standard pipe diameter, and calculating the pipe diameter ratio between the real-time pipe diameter and the standard pipe diameter;
And determining the stenosis rate of the coronary artery blood vessel through the pipe diameter ratio.
2. The method according to claim 1, wherein the generating at least one coronary image combination to be detected from the sequence of coronary images to be detected comprises:
Starting from a first coronary artery image to be detected of the coronary artery image sequence to be detected, selecting adjacent preset number of coronary artery images to be detected as a first coronary artery image combination to be detected;
starting from a second coronary artery image to be detected of the coronary artery image sequence to be detected, selecting adjacent preset number of coronary artery images to be detected as a second coronary artery image combination to be detected;
repeating the operation until each coronary artery image to be detected of the coronary artery image sequence to be detected is traversed.
3. The method of claim 1, wherein the inputting the combination of coronary artery images to be detected into a trained plaque detection model comprises:
Normalizing the spatial coordinate values of the intermediate sequence coronary artery images to be detected in the coronary artery image combination to be detected;
And inputting the spatial coordinate values after the coronary artery image combination to be detected and normalization to a trained plaque detection model.
4. A plaque detection apparatus, comprising:
The image sequence acquisition module is used for acquiring a coronary artery image sequence to be detected;
the image combination generating module is used for generating at least one coronary artery image combination to be detected according to the coronary artery image sequence to be detected, and the coronary artery image combination to be detected comprises at least two coronary artery images to be detected;
The plaque detection module is used for inputting the combination of the coronary artery images to be detected to a trained plaque detection model to obtain an intermediate plaque detection result of each coronary artery image to be detected, wherein the plaque detection model is trained in advance and is used for detecting plaque properties and plaque positions of the coronary artery images to be detected, and the plaque properties comprise: calcified plaque, non-calcified plaque, and mixed plaque;
The final plaque detection result determining module is used for determining a final plaque detection result based on the intermediate plaque detection result of each coronary artery image to be detected; comprising the following steps: the area cross-over ratio determining unit and the plaque position clustering unit;
The area intersection ratio determining unit is used for determining the intersection ratio between plaque areas in each two coronary artery images to be detected;
The plaque position clustering unit is used for clustering the first plaque position in the coronary artery image to be detected, wherein the intersection ratio of the first plaque position and the second plaque position exceeds a preset threshold value, and the final plaque position in the coronary artery image to be detected is obtained; wherein the intermediate plaque detection result includes: the plaque detection method comprises the steps that a plaque detection result comprises a final plaque position, wherein the first plaque position is a two-dimensional plaque position, and the final plaque position is a three-dimensional plaque position;
The final plaque detection result determining module further includes: a plaque number statistics unit and a final plaque property determination unit; the plaque quantity counting unit is used for counting the plaque quantity of each plaque property in the coronary artery image to be detected, wherein the plaque quantity is larger than a preset threshold;
The final plaque property determining unit is used for determining the final property of the plaque according to the plaque number of each type of plaque property; wherein the intermediate plaque detection result includes: each type of property of the plaque, the final plaque detection results including final properties of the plaque;
The image segmentation module is used for carrying out image segmentation on the coronary artery image to be detected according to the final plaque position included in the final plaque detection result so as to extract the coronary artery blood vessel where the plaque is located;
a stenosis rate calculation module for calculating a vessel stenosis rate from the extracted coronary vessels
Wherein the final plaque property determining unit includes:
The plaque sorting subunit is used for sorting the plaque numbers with different properties in a descending order, wherein the sorting order of the plaque numbers with different properties is as follows: calcified plaque, mixed plaque, non-calcified plaque; a final plaque property determining subunit, configured to determine that the final property of the plaque is a mixed plaque if the number of plaques of the two types of properties and the ratio of the total number of the plaques are greater than a preset ratio; if the ratio of the number of the two types of the plaques to the total number of the plaques is smaller than a preset ratio, determining the final property of the plaques to be calcified plaques, wherein the two types of the plaques are mixed plaques and non-calcified plaques;
The stenosis rate calculation module is specifically configured to determine a midline of the coronary artery vessel by a midline extraction technology, and measure a distance from an edge of the coronary artery vessel to the midline, where a caliber of the coronary artery vessel is twice as large as a distance from the edge of the coronary artery vessel to the midline; measuring the real-time pipe diameter of the coronary artery blood vessel in the measurement mode, presetting a standard pipe diameter, and calculating the pipe diameter ratio between the real-time pipe diameter and the standard pipe diameter; and determining the stenosis rate of the coronary artery blood vessel through the pipe diameter ratio.
5. An apparatus, the apparatus comprising:
one or more processors;
A storage means for storing one or more programs;
The one or more programs, when executed by the one or more processors, cause the one or more processors to implement the plaque detection method of any of claims 1-3.
6. A computer-readable storage medium, on which a computer program is stored, characterized in that the program, when being executed by a processor, implements the plaque detection method as claimed in any one of claims 1-3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010693473.XA CN111862038B (en) | 2020-07-17 | 2020-07-17 | Plaque detection method, plaque detection device, plaque detection equipment and plaque detection medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010693473.XA CN111862038B (en) | 2020-07-17 | 2020-07-17 | Plaque detection method, plaque detection device, plaque detection equipment and plaque detection medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111862038A CN111862038A (en) | 2020-10-30 |
| CN111862038B true CN111862038B (en) | 2024-05-14 |
Family
ID=73000986
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010693473.XA Active CN111862038B (en) | 2020-07-17 | 2020-07-17 | Plaque detection method, plaque detection device, plaque detection equipment and plaque detection medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111862038B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113096097A (en) * | 2021-04-13 | 2021-07-09 | 上海商汤智能科技有限公司 | Blood vessel image detection method, detection model training method, related device and equipment |
| CN113034507A (en) * | 2021-05-26 | 2021-06-25 | 四川大学 | CCTA image-based coronary artery three-dimensional segmentation method |
| CN113538471B (en) * | 2021-06-30 | 2023-09-22 | 上海联影医疗科技股份有限公司 | Plaque segmentation method, plaque segmentation device, computer equipment and storage medium |
| CN114049282B (en) * | 2022-01-07 | 2022-05-24 | 浙江大学 | Coronary artery construction method, device, terminal and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109493280A (en) * | 2018-11-02 | 2019-03-19 | 腾讯科技（深圳）有限公司 | Image processing method, device, terminal and storage medium |
| CN110276780A (en) * | 2019-06-17 | 2019-09-24 | 广州织点智能科技有限公司 | A kind of multi-object tracking method, device, electronic equipment and storage medium |
| CN110796186A (en) * | 2019-10-22 | 2020-02-14 | 华中科技大学无锡研究院 | Dry and wet garbage identification and classification method based on improved YOLOv3 network |
| CN111145173A (en) * | 2019-12-31 | 2020-05-12 | 上海联影医疗科技有限公司 | Plaque identification method, device, equipment and medium for coronary angiography image |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10483006B2 (en) * | 2017-05-19 | 2019-11-19 | Siemens Healthcare Gmbh | Learning based methods for personalized assessment, long-term prediction and management of atherosclerosis |
-
2020
- 2020-07-17 CN CN202010693473.XA patent/CN111862038B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109493280A (en) * | 2018-11-02 | 2019-03-19 | 腾讯科技（深圳）有限公司 | Image processing method, device, terminal and storage medium |
| CN110276780A (en) * | 2019-06-17 | 2019-09-24 | 广州织点智能科技有限公司 | A kind of multi-object tracking method, device, electronic equipment and storage medium |
| CN110796186A (en) * | 2019-10-22 | 2020-02-14 | 华中科技大学无锡研究院 | Dry and wet garbage identification and classification method based on improved YOLOv3 network |
| CN111145173A (en) * | 2019-12-31 | 2020-05-12 | 上海联影医疗科技有限公司 | Plaque identification method, device, equipment and medium for coronary angiography image |
Non-Patent Citations (1)
| Title |
|---|
| 改进YOLO的密集小尺度人脸检测方法;邓珍荣;白善今;马富欣;黄文明;;计算机工程与设计(第03期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111862038A (en) | 2020-10-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111862038B (en) | Plaque detection method, plaque detection device, plaque detection equipment and plaque detection medium | |
| US11101033B2 (en) | Medical image aided diagnosis method and system combining image recognition and report editing | |
| US9881372B2 (en) | Method and system for vascular disease detection using recurrent neural networks | |
| CN113781440B (en) | Ultrasonic video focus detection method and device | |
| CN103886576B (en) | A kind of gland tissue signature grey scale detection method and device | |
| CN110858399B (en) | Method and apparatus for providing post-examination images of a virtual tomographic stroke | |
| CN103337096A (en) | Coronary artery CT (computed tomography) contrastographic image calcification point detecting method | |
| KR101697880B1 (en) | Method and device for detecting the volume of atherosclerotic plaque in CT images using adaptive thresholding segmentation | |
| CN114170134A (en) | Stenosis assessment method and device based on intracranial DSA image | |
| KR102206621B1 (en) | Programs and applications for sarcopenia analysis using deep learning algorithms | |
| CN108198174A (en) | A kind of angiocarpy IVOCT and IVUS autoegistration methods and device | |
| CN110349143A (en) | A kind of method, apparatus, equipment and the medium of determining tubular tissue region of interest | |
| Behnami et al. | Automatic detection of patients with a high risk of systolic cardiac failure in echocardiography | |
| CN107194922B (en) | A kind of extracting method of intravascular ultrasound image sequence key frame | |
| Mahfouz et al. | Ultrafast localization of the optic disc using dimensionality reduction of the search space | |
| Tsai et al. | Adaptive segmentation of vessels from coronary angiograms using multi-scale filtering | |
| AU2021106029A4 (en) | Plaque Detection Method, Device, Device and Medium | |
| JP3647970B2 (en) | Region extraction device | |
| EP1447772A1 (en) | A method of lung lobe segmentation and computer system | |
| TWI490790B (en) | Dynamic cardiac imaging analysis and cardiac function assessment system | |
| CN114841914A (en) | Method and device for detecting cystic lesions of pancreas | |
| Sui et al. | A novel method for vessel segmentation and automatic diagnosis of vascular stenosis | |
| CN111028219B (en) | Colon image recognition method and device and related equipment | |
| Shakir et al. | Early detection, segmentation and quantification of coronary artery blockage using efficient image processing technique | |
| US20230401712A1 (en) | Method for determining cardiac coronary artery imaging phase, electronic device, and storage medium |
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 | ||
| CB02 | Change of applicant information |
Address after: 100037 Beijing City, Xicheng District No. 167 North Applicant after: FUWAI HOSPITAL, CHINESE ACADEMY OF MEDICAL SCIENCES Applicant after: Tuxiang Medical Technology Co.,Ltd. Address before: 100037 Beijing City, Xicheng District No. 167 North Applicant before: FUWAI HOSPITAL, CHINESE ACADEMY OF MEDICAL SCIENCES Applicant before: INFERVISION |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |