CN116206162B - Coronary blood flow reserve acquisition method, device and equipment based on contrast image - Google Patents
Coronary blood flow reserve acquisition method, device and equipment based on contrast image Download PDFInfo
- Publication number
- CN116206162B CN116206162B CN202310480472.0A CN202310480472A CN116206162B CN 116206162 B CN116206162 B CN 116206162B CN 202310480472 A CN202310480472 A CN 202310480472A CN 116206162 B CN116206162 B CN 116206162B
- Authority
- CN
- China
- Prior art keywords
- image
- length
- image data
- blood vessel
- target blood
- 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; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
-
- 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
- G06T2207/10016—Video; Image sequence
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Abstract
The application relates to a coronary blood flow reserve obtaining method, device and equipment based on contrast images, which are used for obtaining a resting image group and a congestion image group by screening and classifying an obtained angiography image data set, dividing image data in the resting image group and the congestion image group frame by frame to obtain a binary image of a target blood vessel in each frame image, so as to obtain a length-frame number curve of each image data corresponding to the target blood vessel, correcting the length of the target blood vessel in each image data to obtain a corrected length of the target blood vessel, and finally processing the corrected length in the corresponding length-frame number curve and calculating to obtain the coronary blood flow reserve. By adopting the method, accurate coronary blood flow reserve value can be obtained only by using the coronary angiography image without using a guide wire and a temperature sensor.
Description
Technical Field
The present disclosure relates to the field of medical image processing technologies, and in particular, to a method, an apparatus, and a device for obtaining coronary blood flow reserve based on contrast images.
Background
Coronary flow reserve (Coronary flow reserve, CFR) is an important indicator in the evaluation of coronary functioning, defined as the ratio of the blood flow of the coronary system at its maximum hyperemic state to the blood flow at rest, reflecting the maximum capacity of the coronary system to increase in blood flow from rest to hyperemic state. In current clinical practice, coronary reserves are often measured in the catheter chamber by a "thermal dilution method" which characterizes the blood flow in terms of the time of transport of the blood flow from the coronary orifice to the distal end of the vessel, in particular by a calculation formula cfr=tmn_rest/tmn_hyp, where tmn_rest is the time of transport measured in resting state and tmn_hyp is the time of transport measured in maximum hyperemic state. Notably, in the "thermal dilution method", it is necessary to use a guide wire to deliver the temperature sensor to the distal end of the blood vessel to measure the delivery time Tmn. Specifically, at least three times of bolus injection of normal saline at room temperature into the coronary system of the patient are required before the injection of the hyperemic agent, so that 3 resting state transport time measurement values are obtained, and the average value of the three is the final Tmn_rest. After the injection of the hyperemic agent to induce the coronary system to enter the maximum hyperemic state, the above procedure needs to be repeated to measure the final tmn_hyp, thus at least 6 injections of saline are required in this method, which is more traumatic to the patient.
In summary, although the "thermal dilution method" is the most widely used CFR measurement method in current clinical practice, the method has various problems of long measurement time, high empirical requirements, high cost, low repeatability, high traumatism and the like, and greatly limits the further development of the technology in clinic.
Disclosure of Invention
In view of the foregoing, it is desirable to provide a method, an apparatus, and a device for obtaining coronary flow reserve based on a contrast image, which can accurately obtain the coronary flow reserve without invading a human body with an instrument.
A method of coronary flow reserve acquisition based on contrast imaging, the method comprising:
acquiring an angiographic image dataset of a coronary artery, the angiographic image dataset comprising a plurality of image data;
processing each image data by adopting a pre-trained first deep learning network according to the designated target blood vessel type, outputting a probability value corresponding to each image data as the designated target blood vessel type, and screening each image data according to the probability value to obtain a target blood vessel image set;
classifying each image data in the target blood vessel image set according to the time of injecting the hyperemic agent, taking the image data obtained before injecting the hyperemic agent as a resting image group, and taking the image data obtained after injecting the hyperemic agent as a hyperemic image group;
The image data in the rest image group and the congestion image group are segmented frame by adopting a pre-trained second deep learning network to obtain a binary image of a target blood vessel in each frame of image, and each binary image is processed to obtain a length-frame number curve of the target blood vessel corresponding to each image data;
selecting one image data from the rest image group and the congestion image group according to the probability value output by the first deep learning network as a reference image, determining a reference length and a reference point on a target blood vessel on the reference image, and calibrating the target blood vessel length in other image data according to the reference point to obtain a corresponding calibration length;
acquiring corresponding initial frames and termination frames on a length-frame number curve corresponding to the image data according to the reference length or the calibration length, calculating according to the initial frames, the termination frames and the contrast parameter data to obtain the transportation time of the same section of target blood vessel corresponding to each image data, and correspondingly obtaining the resting state transportation time and the congestion state transportation time;
and calculating according to the resting state transportation time and the hyperemic state transportation time to obtain coronary blood flow reserve.
In one embodiment, the filtering the image data according to the probability value to obtain the target blood vessel image set includes:
and screening out the image data with the probability value higher than a preset probability threshold value to construct the target blood vessel image set, wherein the preset probability threshold value is 0.5-0.7.
In one embodiment, the classifying of the image data in the target blood vessel image set is performed according to the time of injection of the hyperemic agent:
when the number of the screened image data in the resting image group or the congestion image group is less than or equal to 3, reserving all the screened image data;
and when the number of the image data screened into the rest image group or the congestion image group is more than 3, after the probability values output by the first deep learning network are ordered from high to low, the image data corresponding to the first 3 are reserved in the corresponding image group.
In one embodiment, the processing the binary maps to obtain the length-frame number curve of the target blood vessel corresponding to each image data includes:
extracting the central line of the target blood vessel by adopting a refinement algorithm to the binary image, thereby obtaining the length of the blood vessel;
taking the image frame number as an abscissa in each image data, and taking the blood vessel length obtained by each frame of image as a corresponding ordinate to obtain an initial length-frame number curve;
And smoothing, differencing or fitting each initial length-frame curve to obtain final length-frame curves respectively.
In one embodiment, selecting one image data from the rest image group and the congestion image group according to the probability value output by the first deep learning network as a reference image, and determining a reference length and a reference point on a target blood vessel on the reference image includes:
taking image data corresponding to the highest probability value output by the first deep learning network as a reference image;
calculating the maximum length of the blood vessel on a length-frame number curve corresponding to the reference image, and calculating according to the maximum length and preset parameters to obtain the reference length;
and determining an intersection point on the length-frame number curve according to the reference length, and mapping the intersection point to a reference image to obtain the reference point.
In one embodiment, the calibrating the target blood vessel length in the other image data according to the reference point to obtain the corresponding calibration length includes:
and correcting the lengths of the target blood vessels in the rest image group and other image data in the congestion image group according to the reference image and the reference point respectively to obtain a calibration point, so that a target blood vessel with the length consistent with the reference length is determined in the corresponding image data according to the calibration point, and the calibration length is obtained according to the calibration point.
In one embodiment, the correcting the target blood vessel length in the rest image set and the other image data in the congestion image set according to the reference image and the reference point respectively and obtaining a calibration point includes:
taking image data except the reference image as a calibration image, and obtaining a spatial position relation between the reference image and the calibration image according to contrast parameter data of the reference image and the calibration image;
combining the spatial position relation and the position of the datum point on the reference image to calculate and obtain the polar line position of the point on the calibration image;
and the intersection point of the polar line and the target blood vessel in the calibration image is the calibration point.
In one embodiment, the obtaining the corresponding start frame and the end frame on the length-frame number curve corresponding to the image data according to the reference length or the calibration length includes:
searching a point with the length of the first blood vessel not being 0 on the length-frame number curve, wherein the abscissa corresponding to the point is the initial frame;
and starting searching on the length-frame number curve by taking the initial frame as a starting point to obtain a point that the length of the first blood vessel reaches the reference length or the calibration length, wherein the abscissa corresponding to the point is the termination frame.
A contrast image-based coronary flow reserve acquisition device, the device comprising:
the image data acquisition module is used for acquiring an angiography image data set of a coronary artery, wherein the angiography image data set comprises a plurality of image data;
the target blood vessel image set obtaining module is used for processing each image data by adopting a pre-trained first deep learning network according to the appointed target blood vessel type, outputting a probability value corresponding to each image data as the appointed target blood vessel type, and screening each image data according to the probability value to obtain a target blood vessel image set;
the image data classification module is used for classifying each image data in the target blood vessel image set according to the time of injecting the hyperemic agent, taking the image data obtained before the hyperemic agent is injected as a rest image group and the image data obtained after the hyperemic agent is injected as a hyperemic image group;
the length-frame number curve obtaining module of the target blood vessel is used for dividing the image data in the resting image group and the congestion image group frame by adopting a pre-trained second deep learning network to obtain a binary image of the target blood vessel in each frame of image, and processing each binary image to obtain a length-frame number curve of the target blood vessel corresponding to each image data;
The calibration length obtaining module is used for selecting one image data from the rest image group and the congestion image group according to the probability value output by the first deep learning network as a reference image, determining a reference length and a reference point on a target blood vessel on the reference image, and calibrating the target blood vessel length in other image data according to the reference point to obtain a corresponding calibration length;
the target intravascular blood transportation time calculation module is used for acquiring corresponding initial frames and termination frames on a length-frame number curve corresponding to the image data according to the reference length or the calibration length, calculating according to the initial frames, the termination frames and the contrast parameter data to obtain transportation time of the same section of target blood in each image data, and correspondingly obtaining resting state transportation time and congestion state transportation time;
and the coronary blood flow reserve obtaining module is used for calculating according to the resting state transportation time and the congestion state transportation time to obtain the coronary blood flow reserve.
A computer device comprising a memory storing a computer program and a processor which when executing the computer program performs the steps of:
Acquiring an angiographic image dataset of a coronary artery, the angiographic image dataset comprising a plurality of image data;
processing each image data by adopting a pre-trained first deep learning network according to the designated target blood vessel type, outputting a probability value corresponding to each image data as the designated target blood vessel type, and screening each image data according to the probability value to obtain a target blood vessel image set;
classifying each image data in the target blood vessel image set according to the time of injecting the hyperemic agent, taking the image data obtained before injecting the hyperemic agent as a resting image group, and taking the image data obtained after injecting the hyperemic agent as a hyperemic image group;
the image data in the rest image group and the congestion image group are segmented frame by adopting a pre-trained second deep learning network to obtain a binary image of a target blood vessel in each frame of image, and each binary image is processed to obtain a length-frame number curve of the target blood vessel corresponding to each image data;
selecting one image data from the rest image group and the congestion image group according to the probability value output by the first deep learning network as a reference image, determining a reference length and a reference point on a target blood vessel on the reference image, and calibrating the target blood vessel length in other image data according to the reference point to obtain a corresponding calibration length;
Acquiring corresponding initial frames and termination frames on a length-frame number curve corresponding to the image data according to the reference length or the calibration length, calculating according to the initial frames, the termination frames and the contrast parameter data to obtain the transportation time of the same section of target blood vessel corresponding to each image data, and correspondingly obtaining the resting state transportation time and the congestion state transportation time;
and calculating according to the resting state transportation time and the hyperemic state transportation time to obtain coronary blood flow reserve.
A computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of:
acquiring an angiographic image dataset of a coronary artery, the angiographic image dataset comprising a plurality of image data;
processing each image data by adopting a pre-trained first deep learning network according to the designated target blood vessel type, outputting a probability value corresponding to each image data as the designated target blood vessel type, and screening each image data according to the probability value to obtain a target blood vessel image set;
classifying each image data in the target blood vessel image set according to the time of injecting the hyperemic agent, taking the image data obtained before injecting the hyperemic agent as a resting image group, and taking the image data obtained after injecting the hyperemic agent as a hyperemic image group;
The image data in the rest image group and the congestion image group are segmented frame by adopting a pre-trained second deep learning network to obtain a binary image of a target blood vessel in each frame of image, and each binary image is processed to obtain a length-frame number curve of the target blood vessel corresponding to each image data;
selecting one image data from the rest image group and the congestion image group according to the probability value output by the first deep learning network as a reference image, determining a reference length and a reference point on a target blood vessel on the reference image, and calibrating the target blood vessel length in other image data according to the reference point to obtain a corresponding calibration length;
acquiring corresponding initial frames and termination frames on a length-frame number curve corresponding to the image data according to the reference length or the calibration length, calculating according to the initial frames, the termination frames and the contrast parameter data to obtain the transportation time of the same section of target blood vessel corresponding to each image data, and correspondingly obtaining the resting state transportation time and the congestion state transportation time;
and calculating according to the resting state transportation time and the hyperemic state transportation time to obtain coronary blood flow reserve.
According to the coronary blood flow reserve obtaining method, the coronary blood flow reserve obtaining device and the coronary blood flow reserve obtaining equipment based on the contrast images, the obtained angiography image data sets are screened and classified to obtain the rest image set and the congestion image set, the image data in the rest image set and the congestion image set are segmented frame by frame to obtain a binary image of a target blood vessel in each frame of image, so that a length-frame number curve of the target blood vessel corresponding to each image data is obtained, then the length of the target blood vessel in each image data is corrected to obtain the corrected length of the target blood vessel, finally the coronary blood flow reserve is obtained through calculation by processing according to the corrected length in the corresponding length-frame number curve. By adopting the method, accurate coronary blood flow reserve value can be obtained only by using the coronary angiography image without using a guide wire and a temperature sensor.
Drawings
FIG. 1 is a flow chart of a method for obtaining coronary reserves based on contrast images according to one embodiment;
FIG. 2 is a schematic diagram illustrating a result of frame-by-frame segmentation of an image in image data according to an embodiment;
FIG. 3 is a schematic diagram of a length-frame number curve in one embodiment;
FIG. 4 is a schematic diagram of a vessel length calibration in one embodiment;
FIG. 5 is a diagram of the calculation results of a start frame and a stop frame in one embodiment;
FIG. 6 is a block diagram of a coronary flow reserve acquisition device for acquiring contrast-based images in one embodiment;
fig. 7 is an internal structural diagram of a computer device in one embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application.
Aiming at the problems of high operation difficulty, high cost, low repeatability, high traumata and the like of a thermal dilution method adopted in the measurement of coronary blood flow reserve in the prior art, as shown in fig. 1, the coronary blood flow reserve acquisition method based on contrast image is provided, and comprises the following steps:
step S100, acquiring an angiography image data set of a coronary artery, wherein the angiography image data set comprises a plurality of image data;
step S110, processing each image data by adopting a pre-trained first deep learning network according to the designated target blood vessel type, outputting a probability value corresponding to each image data as the designated target blood vessel type, and screening each image data according to the probability value to obtain a target blood vessel image set;
Step S120, classifying each image data in the target blood vessel image set according to the time of injecting the hyperemic agent, taking the image data obtained before injecting the hyperemic agent as a rest image group, and taking the image data obtained after injecting the hyperemic agent as a hyperemic image group;
step S130, the image data in the rest image group and the congestion image group are segmented frame by adopting a pre-trained second deep learning network to obtain a binary image of a target blood vessel in each frame of image, and each binary image is processed to obtain a length-frame number curve of the target blood vessel corresponding to each image data;
step S140, selecting one image data from the rest image group and the congestion image group according to the probability value output by the first deep learning network as a reference image, determining a reference length and a reference point on a target blood vessel on the reference image, and calibrating the target blood vessel length in other image data according to the reference point to obtain a corresponding calibration length;
step S150, acquiring corresponding initial frames and termination frames on a length-frame number curve corresponding to the image data according to the reference length or the calibration length, calculating according to the initial frames, the termination frames and the contrast parameter data to obtain the transportation time of the same section of target blood vessel corresponding to each image data, and correspondingly obtaining the resting state transportation time and the congestion state transportation time;
Step S160, calculating according to the resting state transportation time and the hyperemic state transportation time to obtain coronary blood flow reserve.
In the method, accurate coronary blood flow reserve data can be obtained by only processing angiographic images of coronary arteries, so as to overcome the problems in the prior art.
In step S100, angiographic image data of a coronary artery is obtained by a digital silhouette angiography (ICA) technique, and each image data includes a plurality of angiographic images arranged in time sequence.
In step S110, the specified target vessel type is one of the anterior descending, the circumflex, and the right coronary vessels. When the first deep learning network trained in advance is adopted to process each image data, each image data and the corresponding contrast angle are input into the deep learning network, and the probability value of the corresponding image data belonging to the target blood vessel type is output. When each image data is screened according to the probability value, the image data with the probability value higher than a preset probability threshold value is screened out to construct a target blood vessel image set, wherein in one embodiment, the preset probability threshold value is 0.5-0.7.
In step S120, all image data in the target blood vessel image set are divided into a rest image group and a congestion image group according to the time when the operator injects the congestion agent during the contrast operation. And further, when the number of the image data in the screened resting image group or the hyperemia image group is less than or equal to 3, all the screened image data are reserved. And when the number of the image data screened into the resting image group or the congestion image group is more than 3, the image data corresponding to the first 3 are reserved in the corresponding image group after the probability values output by the first deep learning network are ordered from high to low. That is, in this step, the image data in the target blood vessel image set is further subjected to further screening and classification.
In step S130, the image data is segmented by using a pre-trained second deep learning network to obtain a binary image corresponding to each frame of image in each image data, as shown in fig. 2.
In this embodiment, the processing of each binary image to obtain the length-frame number curve of the target blood vessel corresponding to each image data includes: extracting the central line of the target blood vessel by adopting a refinement algorithm to the binary image to obtain the length of the blood vessel, taking the number of image frames as the abscissa in each image data, taking the length of the blood vessel obtained by each frame of image as the corresponding ordinate to obtain an initial length-frame number curve, and carrying out smoothing, difference or fitting treatment on each initial length-frame number curve to obtain a final length-frame number curve respectively, as shown in figure 3.
Because coronary angiography is a 2D imaging mode, a certain projection reduction problem exists, namely, the same target vessel can obtain different 2D projection lengths in the contrast images of different contrast angles. On the other hand, when calculating coronary reserves by the transit time, it is necessary to ensure that all resting and engorged transit times are measured based on the same length of the same segment of blood vessel. Therefore, in order to ensure accurate measurement of coronary blood flow reserve, in the method, the resting image group and the hyperemic image group are calibrated for target blood vessel length, so as to unify the calculated blood vessel lengths of the coronary blood flow reserve, and it should be noted that the unification refers to spatial unification, that is, the actual length of the target blood vessel, and not to image unification, that is, different 2D blood vessel lengths may be obtained from different contrast angle images.
When correcting the target blood vessel length, firstly selecting one image data from the rest image group and the congestion image group as a reference image, and determining the reference length and the reference point on the target blood vessel on the reference image. And then correcting other image data according to the reference image.
In step S140, selecting one image data from the rest image group and the congestion image group as the reference image includes: and taking the image data corresponding to the highest probability value output by the first deep learning network as a reference image, calculating the maximum length of the blood vessel on a length-frame curve corresponding to the reference image, calculating according to the maximum length and preset parameters to obtain a reference length, determining an intersection point on the length-frame curve according to the reference length, and mapping the intersection point to the reference image to obtain the reference point.
Specifically, in all the rest group images and the congestion group images, the image with the highest probability value is selected as a reference image for length calibration based on the probability value output by the first deep learning network. On a length-frame number curve L (frame) corresponding to the reference image, a maximum value Lmax of the target blood vessel length is calculated, and 0.7×lmax is taken as the reference length of the reference image. The first intersection point of the straight line l=0.7×lmax and the curve l=l (frame) is obtained, and the point is mapped back to the reference image to obtain a reference point, fig. 4 is a schematic diagram of the calibration of the blood vessel length in one embodiment, the left side image is the reference image, and the reference point found on the target blood vessel by the above steps, where the reference point is within the scope of the left side image.
And then, calibrating the target blood vessel length in other image data according to the reference point to obtain a corresponding calibration length. Specifically, calibrating the target blood vessel length in other image data according to the reference point to obtain a corresponding calibration length includes: and correcting the lengths of the target blood vessels in the rest image group and other image data in the congestion image group according to the reference image and the reference point respectively to obtain a calibration point, so that a target blood vessel with the length consistent with the reference length is determined in the corresponding image data according to the calibration point, and the calibration length is obtained according to the calibration point.
Further, correcting the target blood vessel length in the rest image group and the other image data in the congestion image group according to the reference image and the reference point respectively and obtaining a calibration point comprises: and taking image data except the reference image as a calibration image, obtaining a spatial position relation between the reference image and the calibration image according to contrast parameter data of the reference image and the calibration image, and calculating the polar line position of the point on the calibration image by combining the spatial position relation and the position of the reference point on the reference image, wherein the intersection point of the polar line in the calibration image and the target blood vessel is the calibration point. As shown in fig. 4, the right-hand graph is a calibration image in which a calibration point is found on the target vessel according to the reference point and the above steps, the calibration point being within the range enclosed in the right-hand graph.
Specifically, the rest state and hyperemia state images are selected one by one to serve as calibration images, the spatial position relation between the rest state and hyperemia state images can be calculated based on contrast parameter information of the reference images and the calibration images, and then the polar line position of the point on the calibration images can be calculated by combining the reference point positions, so that the intersection point of the polar line and blood vessels on the calibration images is the calibration point. On the calibration image, the length of the blood vessel from the blood vessel starting point to the calibration point is the calibration length of the calibration image.
In step S150, obtaining a corresponding start frame and a corresponding end frame on a length-frame number curve corresponding to the image data according to the calibrated blood vessel length in each image data includes: searching a point with the length of the first blood vessel not being 0 on the length-frame number curve, wherein the abscissa corresponding to the point is a starting frame, starting searching on the length-frame number curve by taking the starting frame as a starting point, obtaining a point with the length of the first blood vessel reaching a reference length (the reference length corresponds to a reference image) or a calibration length (the calibration length corresponds to a calibration image), and taking the abscissa corresponding to the point as a termination frame.
Specifically, for any resting group or hyperemic groupFirst, a point with the length of the first blood vessel not being 0 is searched on the corresponding length-frame number curve, the point represents that the contrast agent just flows into the target blood vessel, and the abscissa corresponding to the point is the initial frame . From the start frame onwards, a point is searched for where the first vessel length reaches the reference or calibration length, which point indicates that the contrast agent has just flowed to the reference or calibration point, the abscissa corresponding to which point is the stop frame +.>. Fig. 5 is a schematic diagram of calculation results of a start frame and a stop frame in an embodiment.
In this embodiment, the following formula may be adopted for calculating the transport time of the same segment of target blood vessel corresponding to each image data according to the start frame, the end frame and the contrast parameter data:
（1）
in the case of the formula (1),the contrast DSA image is represented by its own parameter time resolution, which is defined as the number of imaging frames per second of image.
Based on the time of transportationThe average blood flow velocity in the target blood vessel can be calculated>Flow->The specific calculation mode is as follows:
（2）
（3）
in formulas (2) and (3),representing a preset vascular length, < >>Represents the average cross-sectional area of the target vessel segment. Preferably, the method comprises the steps of,the value is 75 mm.
Corresponding transportation time can be obtained after the image data in the resting image group and the congestion image group are processed one by one, so that the resting transportation time is obtained by calculating the average transportation time according to all the transportation time obtained by the resting image group,
And similarly, calculating average transportation time according to all transportation time obtained by the hyperemic image group to obtain the hyperemic transportation time. In other embodiments, to fully simulate the maximum congestion state, the minimum transit time is selected from the congestion image group calculation results as the final congestion state transit time.
Finally, in step S160, a calculation formula of coronary blood flow reserve is used to calculate the value of coronary blood flow reserve. Wherein, the calculation formula is: cfr=tmn_rest/tmn_hyp, where tmn_rest is the final resting state transport time and tmn_hyp is the final hyperemic state transport time.
In the coronary blood flow reserve acquisition method based on the contrast image, only the coronary contrast image is needed to be imported, and a guide wire and a temperature sensor are not needed to be used, so that the room temperature physiological saline is not needed to be injected into the coronary blood vessel of the patient for multiple times. Compared with the traditional thermal dilution method, the method reduces the trauma to the patient, reduces the operation difficulty and saves the operation cost.
It should be understood that, although the steps in the flowchart of fig. 1 are shown in sequence as indicated by the arrows, the steps are not necessarily performed in sequence as indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in fig. 1 may include multiple sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, nor do the order in which the sub-steps or stages are performed necessarily performed in sequence, but may be performed alternately or alternately with at least a portion of other steps or sub-steps of other steps.
In one embodiment, as shown in fig. 6, there is provided a coronary flow reserve acquisition device based on a contrast image, comprising: the image data acquisition module 200, the target blood vessel image set obtaining module 210, the image data classification module 220, the target blood vessel length-frame number curve obtaining module 230, the calibration length obtaining module 240, the target blood vessel internal blood transportation time calculating module 250 and the coronary blood flow reserve obtaining module 260, wherein:
an image data acquisition module 200, configured to acquire an angiographic image dataset of a coronary artery, where the angiographic image dataset includes a plurality of image data;
the target blood vessel image set obtaining module 210 is configured to process each image data by using a first deep learning network trained in advance according to a specified target blood vessel type, output a probability value corresponding to each image data as the specified target blood vessel type, and screen each image data according to the probability value to obtain a target blood vessel image set;
the image data classification module 220 is configured to classify each image data in the target blood vessel image set according to the time of injecting the hyperemic agent, take the image data obtained before injecting the hyperemic agent as a rest image group, and take the image data obtained after injecting the hyperemic agent as a hyperemic image group;
A target vessel length-frame number curve obtaining module 230, configured to segment image data in the rest image group and the congestion image group frame by using a pre-trained second deep learning network to obtain a binary image of the target vessel in each frame of image, and process each binary image to obtain a length-frame number curve of the target vessel corresponding to each image data;
the calibration length obtaining module 240 is configured to select one image data from the rest image group and the congestion image group according to the probability value output by the first deep learning network as a reference image, determine a reference length and a reference point on a target blood vessel on the reference image, and calibrate the target blood vessel length in other image data according to the reference point to obtain a corresponding calibration length;
the target intravascular blood transport time calculation module 250 is configured to obtain a corresponding start frame and a corresponding end frame on a length-frame data curve corresponding to the image data according to the reference length or the calibration length, calculate according to the start frame, the end frame and the contrast parameter data to obtain transport time of a target blood vessel corresponding to a same segment in each image data, and obtain a resting state transport time and a congestion state transport time accordingly;
The coronary blood flow reserve obtaining module 260 is configured to calculate a coronary blood flow reserve according to the resting state transportation time and the congestion state transportation time.
For specific limitations on the contrast image-based coronary flow reserve acquisition device, reference may be made to the above limitations on the contrast image-based coronary flow reserve acquisition method, and no further description is given here. The above-mentioned coronary blood flow reserve acquiring device based on the contrast image may be implemented in whole or in part by software, hardware or a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
In one embodiment, a computer device is provided, which may be a terminal, and the internal structure of which may be as shown in fig. 7. The computer device includes a processor, a memory, a network interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a method for obtaining coronary flow reserve based on contrast images. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, can also be keys, a track ball or a touch pad arranged on the shell of the computer equipment, and can also be an external keyboard, a touch pad or a mouse and the like.
It will be appreciated by those skilled in the art that the structure shown in fig. 7 is merely a block diagram of some of the structures associated with the present application and is not limiting of the computer device to which the present application may be applied, and that a particular computer device may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, a computer device is provided comprising a memory and a processor, the memory having stored therein a computer program, the processor when executing the computer program performing the steps of:
acquiring an angiographic image dataset of a coronary artery, the angiographic image dataset comprising a plurality of image data;
processing each image data by adopting a pre-trained first deep learning network according to the designated target blood vessel type, outputting a probability value corresponding to each image data as the designated target blood vessel type, and screening each image data according to the probability value to obtain a target blood vessel image set;
classifying each image data in the target blood vessel image set according to the time of injecting the hyperemic agent, taking the image data obtained before injecting the hyperemic agent as a resting image group, and taking the image data obtained after injecting the hyperemic agent as a hyperemic image group;
The image data in the rest image group and the congestion image group are segmented frame by adopting a pre-trained second deep learning network to obtain a binary image of a target blood vessel in each frame of image, and each binary image is processed to obtain a length-frame number curve of the target blood vessel corresponding to each image data;
selecting one image data from the rest image group and the congestion image group according to the probability value output by the first deep learning network as a reference image, determining a reference length and a reference point on a target blood vessel on the reference image, and calibrating the target blood vessel length in other image data according to the reference point to obtain a corresponding calibration length;
acquiring corresponding initial frames and termination frames on a length-frame number curve corresponding to the image data according to the reference length or the calibration length, calculating according to the initial frames, the termination frames and the contrast parameter data to obtain the transportation time of the same section of target blood vessel corresponding to each image data, and correspondingly obtaining the resting state transportation time and the congestion state transportation time;
and calculating according to the resting state transportation time and the hyperemic state transportation time to obtain coronary blood flow reserve.
In one embodiment, a computer readable storage medium is provided having a computer program stored thereon, which when executed by a processor, performs the steps of:
acquiring an angiographic image dataset of a coronary artery, the angiographic image dataset comprising a plurality of image data;
processing each image data by adopting a pre-trained first deep learning network according to the designated target blood vessel type, outputting a probability value corresponding to each image data as the designated target blood vessel type, and screening each image data according to the probability value to obtain a target blood vessel image set;
classifying each image data in the target blood vessel image set according to the time of injecting the hyperemic agent, taking the image data obtained before injecting the hyperemic agent as a resting image group, and taking the image data obtained after injecting the hyperemic agent as a hyperemic image group;
the image data in the rest image group and the congestion image group are segmented frame by adopting a pre-trained second deep learning network to obtain a binary image of a target blood vessel in each frame of image, and each binary image is processed to obtain a length-frame number curve of the target blood vessel corresponding to each image data;
Selecting one image data from the rest image group and the congestion image group according to the probability value output by the first deep learning network as a reference image, determining a reference length and a reference point on a target blood vessel on the reference image, and calibrating the target blood vessel length in other image data according to the reference point to obtain a corresponding calibration length;
acquiring corresponding initial frames and termination frames on a length-frame number curve corresponding to the image data according to the reference length or the calibration length, calculating according to the initial frames, the termination frames and the contrast parameter data to obtain the transportation time of the same section of target blood vessel corresponding to each image data, and correspondingly obtaining the resting state transportation time and the congestion state transportation time;
and calculating according to the resting state transportation time and the hyperemic state transportation time to obtain coronary blood flow reserve.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the various embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), memory bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The above examples merely represent a few embodiments of the present application, which are described in more detail and are not to be construed as limiting the scope of the invention. It should be noted that it would be apparent to those skilled in the art that various modifications and improvements could be made without departing from the spirit of the present application, which would be within the scope of the present application. Accordingly, the scope of protection of the present application is to be determined by the claims appended hereto.
Claims (9)
1. A method for obtaining coronary flow reserve based on contrast imaging, the method comprising:
acquiring an angiographic image dataset of a coronary artery, the angiographic image dataset comprising a plurality of image data;
processing each image data by adopting a pre-trained first deep learning network according to the designated target blood vessel type, outputting a probability value corresponding to each image data as the designated target blood vessel type, and screening each image data according to the probability value to obtain a target blood vessel image set;
Classifying each image data in the target blood vessel image set according to the time of injecting the hyperemic agent, taking the image data obtained before injecting the hyperemic agent as a resting image group, and taking the image data obtained after injecting the hyperemic agent as a hyperemic image group;
the image data in the rest image group and the congestion image group are segmented frame by adopting a pre-trained second deep learning network to obtain a binary image of a target blood vessel in each frame of image, and each binary image is processed to obtain a length-frame number curve of the target blood vessel corresponding to each image data;
selecting image data from the rest image group and the congestion image group according to the probability value output by the first deep learning network as a reference image, calculating the maximum length of a target blood vessel on a length-frame curve corresponding to the reference image, determining a reference length according to the maximum length, determining an intersection point on a corresponding length-frame curve according to the reference length, and mapping the intersection point to the reference image to determine a reference point;
acquiring the polar line position of the point on other image data by combining the position of the datum point on the datum image through the spatial position relation between the datum image and other image data, so as to determine a calibration point according to the intersection point of the polar line position and the target blood vessel in the corresponding image data, determine a section of target blood vessel with the length consistent with the datum length in the corresponding image data according to each calibration point, and obtain a calibration length according to the calibration point;
Acquiring corresponding initial frames and termination frames on a length-frame number curve corresponding to the image data according to the reference length or the calibration length, calculating according to the initial frames, the termination frames and the contrast parameter data to obtain the transportation time of the same section of target blood vessel corresponding to each image data, and correspondingly obtaining the resting state transportation time and the congestion state transportation time;
and calculating according to the resting state transportation time and the hyperemic state transportation time to obtain coronary blood flow reserve.
2. The method of claim 1, wherein the screening each image data according to the probability value to obtain a target blood vessel image set comprises:
and screening out the image data with the probability value higher than a preset probability threshold value to construct the target blood vessel image set, wherein the preset probability threshold value is 0.5-0.7.
3. The coronary flow reserve acquisition method according to claim 2, wherein, when classifying the image data in the target blood vessel image set according to the time of injection of the hyperemic agent:
when the number of the screened image data in the resting image group or the congestion image group is less than or equal to 3, reserving all the screened image data;
And when the number of the image data screened into the rest image group or the congestion image group is more than 3, after the probability values output by the first deep learning network are ordered from high to low, the image data corresponding to the first 3 are reserved in the corresponding image group.
4. The coronary flow reserve acquisition method as recited in claim 3, wherein said processing each of said binary maps to obtain a length-frame number curve of a target vessel corresponding to each of said image data includes:
extracting the central line of the target blood vessel by adopting a refinement algorithm to the binary image, thereby obtaining the length of the blood vessel;
taking the image frame number as an abscissa in each image data, and taking the blood vessel length obtained by each frame of image as a corresponding ordinate to obtain an initial length-frame number curve;
and smoothing, differencing or fitting each initial length-frame curve to obtain final length-frame curves respectively.
5. The coronary flow reserve acquisition method according to claim 4, wherein selecting one image data from the rest image group and the congestion image group according to the probability value output by the first deep learning network as a reference image, calculating a maximum length of a target blood vessel on a length-frame curve corresponding to the reference image, determining a reference length according to the maximum length, and determining a reference point according to an intersection of the reference length and the corresponding length-frame curve comprises:
Taking image data corresponding to the highest probability value output by the first deep learning network as a reference image;
calculating the maximum length of the blood vessel on a length-frame number curve corresponding to the reference image, and calculating according to the maximum length and preset parameters to obtain the reference length;
and determining an intersection point on the length-frame number curve according to the reference length, and mapping the intersection point to a reference image to obtain the reference point.
6. The method according to claim 5, wherein the determining the calibration point in the target vessel in the other image data by acquiring the epipolar line position in combination with the position of the reference point on the reference image based on the spatial positional relationship between the reference image and the other image data comprises:
taking image data except the reference image as a calibration image, and obtaining a spatial position relation between the reference image and the calibration image according to contrast parameter data of the reference image and the calibration image;
combining the spatial position relation and the position of the datum point on the reference image to calculate and obtain the polar line position of the point on the calibration image;
and the intersection point of the polar line and the target blood vessel in the calibration image is the calibration point.
7. The method according to claim 6, wherein the acquiring the corresponding start frame and end frame on the length-frame number curve corresponding to the image data according to the reference length or the calibration length comprises:
searching a point with the length of the first blood vessel not being 0 on the length-frame number curve, wherein the abscissa corresponding to the point is the initial frame;
and starting searching on the length-frame number curve by taking the initial frame as a starting point to obtain a point that the length of the first blood vessel reaches the reference length or the calibration length, wherein the abscissa corresponding to the point is the termination frame.
8. A contrast image-based coronary flow reserve acquisition device, the device comprising:
the image data acquisition module is used for acquiring an angiography image data set of a coronary artery, wherein the angiography image data set comprises a plurality of image data;
the target blood vessel image set obtaining module is used for processing each image data by adopting a pre-trained first deep learning network according to the appointed target blood vessel type, outputting a probability value corresponding to each image data as the appointed target blood vessel type, and screening each image data according to the probability value to obtain a target blood vessel image set;
The image data classification module is used for classifying each image data in the target blood vessel image set according to the time of injecting the hyperemic agent, taking the image data obtained before the hyperemic agent is injected as a rest image group and the image data obtained after the hyperemic agent is injected as a hyperemic image group;
the length-frame number curve obtaining module of the target blood vessel is used for dividing the image data in the resting image group and the congestion image group frame by adopting a pre-trained second deep learning network to obtain a binary image of the target blood vessel in each frame of image, and processing each binary image to obtain a length-frame number curve of the target blood vessel corresponding to each image data;
the reference length and reference point determining module is used for selecting one image data from the rest image group and the congestion image group according to the probability value output by the first deep learning network as a reference image, calculating the maximum length of a target blood vessel on a length-frame curve corresponding to the reference image, determining a reference length according to the maximum length, determining an intersection point on the corresponding length-frame curve according to the reference length, and mapping the intersection point to the reference image to determine a reference point;
The calibration length obtaining module is used for obtaining the polar line position of the point on other image data by combining the position of the datum point on the datum image through the spatial position relation between the datum image and other image data, so as to determine calibration points according to the intersection points in the target blood vessels in the image data corresponding to the polar line position, determine a section of target blood vessel with the length consistent with the datum length in the corresponding image data according to each calibration point, and obtain the calibration length according to the calibration points;
the target intravascular blood transportation time calculation module is used for acquiring corresponding initial frames and termination frames on a length-frame number curve corresponding to the image data according to the reference length or the calibration length, calculating according to the initial frames, the termination frames and the contrast parameter data to obtain transportation time of the same section of target blood in each image data, and correspondingly obtaining resting state transportation time and congestion state transportation time;
and the coronary blood flow reserve obtaining module is used for calculating according to the resting state transportation time and the congestion state transportation time to obtain the coronary blood flow reserve.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1 to 7 when the computer program is executed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310480472.0A CN116206162B (en) | 2023-04-28 | 2023-04-28 | Coronary blood flow reserve acquisition method, device and equipment based on contrast image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310480472.0A CN116206162B (en) | 2023-04-28 | 2023-04-28 | Coronary blood flow reserve acquisition method, device and equipment based on contrast image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116206162A CN116206162A (en) | 2023-06-02 |
| CN116206162B true CN116206162B (en) | 2023-08-01 |
Family
ID=86509752
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310480472.0A Active CN116206162B (en) | 2023-04-28 | 2023-04-28 | Coronary blood flow reserve acquisition method, device and equipment based on contrast image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116206162B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112971818A (en) * | 2021-01-28 | 2021-06-18 | 杭州脉流科技有限公司 | Method and device for rapidly acquiring microcirculation resistance index, computer equipment and storage medium |
| CN115761132A (en) * | 2022-11-23 | 2023-03-07 | 杭州脉流科技有限公司 | Method and device for automatically reconstructing three-dimensional model of coronary artery |
Family Cites Families (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9702762B2 (en) * | 2013-03-15 | 2017-07-11 | Lightlab Imaging, Inc. | Calibration and image processing devices, methods, and systems |
| US9700219B2 (en) * | 2013-10-17 | 2017-07-11 | Siemens Healthcare Gmbh | Method and system for machine learning based assessment of fractional flow reserve |
| EP3453326A1 (en) * | 2017-09-08 | 2019-03-13 | Koninklijke Philips N.V. | Registration and comparison of measured and simulated intracoronary pullback curves |
| CN108198210B (en) * | 2017-12-28 | 2021-12-21 | 中科微光医疗研究中心(西安)有限公司 | OCT image judging method and system based on mobile equipment |
| CN108186038B (en) * | 2018-02-11 | 2020-11-17 | 杭州脉流科技有限公司 | System for calculating coronary blood flow reserve fraction based on arteriography image |
| CN110786841B (en) * | 2019-11-04 | 2021-05-25 | 苏州润迈德医疗科技有限公司 | Method and device for adjusting maximum hyperemia state flow rate based on microcirculation resistance index |
| CN111161342B (en) * | 2019-12-09 | 2023-08-29 | 杭州脉流科技有限公司 | Method, apparatus, device, system and readable storage medium for obtaining fractional flow reserve based on coronary angiography image |
| CN113487588B (en) * | 2020-03-02 | 2024-03-22 | 上海博动医疗科技股份有限公司 | Coronary artery blood flow velocity calculation method and device and electronic equipment |
| CN111627003B (en) * | 2020-05-25 | 2022-11-29 | 上海杏脉信息科技有限公司 | Coronary blood flow reserve calculating device |
| CN113837985B (en) * | 2020-06-24 | 2023-11-07 | 上海博动医疗科技股份有限公司 | Training method and device for angiographic image processing, automatic processing method and device |
| CN114004793A (en) * | 2021-10-18 | 2022-02-01 | 深圳睿心智能医疗科技有限公司 | Blood flow characteristic acquisition method and device, electronic equipment and storage medium |
| CN115619690A (en) * | 2022-10-12 | 2023-01-17 | 深圳锐视智芯科技有限公司 | Data processing method and related device |
-
2023
- 2023-04-28 CN CN202310480472.0A patent/CN116206162B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112971818A (en) * | 2021-01-28 | 2021-06-18 | 杭州脉流科技有限公司 | Method and device for rapidly acquiring microcirculation resistance index, computer equipment and storage medium |
| CN115761132A (en) * | 2022-11-23 | 2023-03-07 | 杭州脉流科技有限公司 | Method and device for automatically reconstructing three-dimensional model of coronary artery |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116206162A (en) | 2023-06-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111161342B (en) | Method, apparatus, device, system and readable storage medium for obtaining fractional flow reserve based on coronary angiography image | |
| CN111680447B (en) | Blood flow characteristic prediction method, device, computer equipment and storage medium | |
| US7639855B2 (en) | Medical image processing apparatus, and medical image processing method | |
| US20050010100A1 (en) | Method and apparatus for automatic detection of anomalies in vessel structures | |
| CN112971818B (en) | Method and device for acquiring microcirculation resistance index, computer equipment and storage medium | |
| CN113850755A (en) | Method, apparatus, system and computer storage medium for acquiring brain CT perfusion imaging parameter map | |
| CN112435212A (en) | Brain focus region volume obtaining method and device based on deep learning, computer equipment and storage medium | |
| CN110751187A (en) | Training method of abnormal area image generation network and related product | |
| CN116206162B (en) | Coronary blood flow reserve acquisition method, device and equipment based on contrast image | |
| CN114886390A (en) | Method, device, storage medium and program product for determining coronary flow reserve, fractional flow reserve, and microcirculation resistance index | |
| CN116616804B (en) | Method, device, equipment and storage medium for acquiring intracranial arterial stenosis evaluation parameters | |
| CN113995388B (en) | Fractional flow reserve calculation method and device, electronic equipment and readable storage medium | |
| CN109767468A (en) | Internal organs volume detection method and device | |
| US20210219850A1 (en) | Providing a blood flow parameter set for a vascular malformation | |
| CN116205917B (en) | Method, device, computer device and storage medium for obtaining coronary blood flow reserve | |
| CN116704149B (en) | Method and device for obtaining intracranial arterial stenosis parameters | |
| CN117036530B (en) | Cross-modal data-based coronary artery fractional flow reserve prediction method and device | |
| CN116681704B (en) | Intracranial vascular blood flow obtaining method, computer device and readable storage medium | |
| CN116649996B (en) | Method and device for obtaining intracranial arterial stenosis hemodynamic parameters | |
| CN116649995B (en) | Method and device for acquiring hemodynamic parameters based on intracranial medical image | |
| CN117830513A (en) | Method, device and equipment for generating three-dimensional vascular model of intracranial region of interest | |
| EP4310773A1 (en) | Ultrasound data processing | |
| CN114399596A (en) | Method, apparatus, system and computer storage medium for microcatheter shaping | |
| CN115661082A (en) | Coronary vessel segmentation method, apparatus, device, storage medium and program product | |
| CN116030071A (en) | Vascular key point position determination method, vascular key point position determination device, vascular key point position determination computer equipment and vascular key point position 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |