CN116849691A - Method, equipment and storage medium for automatically identifying global optimal phase of cardiac CT imaging - Google Patents
Method, equipment and storage medium for automatically identifying global optimal phase of cardiac CT imaging Download PDFInfo
- Publication number
- CN116849691A CN116849691A CN202311009213.6A CN202311009213A CN116849691A CN 116849691 A CN116849691 A CN 116849691A CN 202311009213 A CN202311009213 A CN 202311009213A CN 116849691 A CN116849691 A CN 116849691A
- Authority
- CN
- China
- Prior art keywords
- phases
- phase
- image
- registration
- cardiac
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 40
- 230000000747 cardiac effect Effects 0.000 title claims abstract description 37
- 238000013170 computed tomography imaging Methods 0.000 title claims abstract description 17
- 239000013598 vector Substances 0.000 claims abstract description 19
- 239000011159 matrix material Substances 0.000 claims description 28
- 230000006870 function Effects 0.000 claims description 11
- 238000013473 artificial intelligence Methods 0.000 claims description 5
- 238000002591 computed tomography Methods 0.000 claims description 5
- 238000004590 computer program Methods 0.000 claims description 3
- 238000000354 decomposition reaction Methods 0.000 claims description 2
- 230000009466 transformation Effects 0.000 claims description 2
- 238000005516 engineering process Methods 0.000 abstract description 3
- 238000002059 diagnostic imaging Methods 0.000 abstract description 2
- 210000004204 blood vessel Anatomy 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 230000006978 adaptation Effects 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000003759 clinical diagnosis Methods 0.000 description 1
- 239000002872 contrast media Substances 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000003205 diastolic effect Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 210000003516 pericardium Anatomy 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus for radiation diagnosis, e.g. combined with radiation therapy equipment
- A61B6/02—Devices for diagnosis sequentially in different planes; Stereoscopic radiation diagnosis
- A61B6/03—Computerised tomographs
- A61B6/032—Transmission computed tomography [CT]
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus for radiation diagnosis, e.g. combined with radiation therapy equipment
- A61B6/50—Clinical applications
- A61B6/503—Clinical applications involving diagnosis of heart
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus for radiation diagnosis, e.g. combined with radiation therapy equipment
- A61B6/52—Devices using data or image processing specially adapted for radiation diagnosis
- A61B6/5205—Devices using data or image processing specially adapted for radiation diagnosis involving processing of raw data to produce diagnostic data
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus for radiation diagnosis, e.g. combined with radiation therapy equipment
- A61B6/52—Devices using data or image processing specially adapted for radiation diagnosis
- A61B6/5211—Devices using data or image processing specially adapted for radiation diagnosis involving processing of medical diagnostic data
- A61B6/5217—Devices using data or image processing specially adapted for radiation diagnosis involving processing of medical diagnostic data extracting a diagnostic or physiological parameter from medical diagnostic data
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus for radiation diagnosis, e.g. combined with radiation therapy equipment
- A61B6/52—Devices using data or image processing specially adapted for radiation diagnosis
- A61B6/5258—Devices using data or image processing specially adapted for radiation diagnosis involving detection or reduction of artifacts or noise
- A61B6/5264—Devices using data or image processing specially adapted for radiation diagnosis involving detection or reduction of artifacts or noise due to motion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/003—Reconstruction from projections, e.g. tomography
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/32—Determination of transform parameters for the alignment of images, i.e. image registration using correlation-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
- G06T7/337—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods involving reference images or patches
-
- 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
- G06T2207/10081—Computed x-ray tomography [CT]
-
- 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
- G06T2207/20084—Artificial neural networks [ANN]
-
- 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
- G06T2207/20212—Image combination
- G06T2207/20221—Image fusion; Image merging
-
- 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
- G06T2207/30048—Heart; Cardiac
Abstract
The invention relates to the field of medical imaging, and discloses a method, equipment and a storage medium for automatically identifying global optimal phases of cardiac CT imaging, wherein the method comprises the following steps: selecting a plurality of phases at intervals in a cardiac cycle, and reconstructing according to scanning data corresponding to the selected phases to obtain a corresponding reconstructed image; registering the adjacent phase images to obtain deformation fields of each pixel of the adjacent phases; according to the deformation field and the time interval of the adjacent phases, the heart motion speed of each pixel at the corresponding moment of the selected phases can be obtained; calculating the average value of the heart motion velocity vector modes of all pixels as the average motion amplitude at the moment; and traversing all phases, wherein the time with the minimum mean value is the optimal phase time. The method provided by the invention determines the deformation fields of the heart motion X, Y and the Z direction through an image registration technology, can quantitatively describe the heart motion, and has small influence of image noise on registration accuracy.
Description
Technical Field
The invention relates to the field of medical imaging, in particular to a method, equipment and a storage medium for automatically identifying global optimal phases of cardiac CT imaging.
Background
In the prior art, the CT scan can acquire a data source of the time period in the heart cycle by utilizing the characteristic of relatively gentle diastole movement in the heart cycle, and acquire three-dimensional reconstruction data of relatively stable heart scan, so as to improve the imaging quality when the CT scan is performed on the heart. At present, after image reconstruction is performed, a doctor usually finds that motion artifacts in an image are serious, and then changes a reconstruction phase to acquire images of other scan intervals according to the performance of the motion artifacts, and then finds a more proper reconstruction phase to serve as a final clinical diagnosis image. Because of the complexity of motion artifacts, conventional manual methods require traversing different phases, and therefore require much reconstruction and processing time, which is difficult to effectively improve efficiency. Traditional manual methods require traversing different phases and therefore require much reconstruction and processing time, as well as increasing the physician workload. However, other existing techniques require registering the blood vessel, and because the blood vessel is very thin, the techniques have high requirements on the extraction accuracy of the blood vessel.
Existing global optimal phase techniques typically reconstruct images of all phases in the range of 0% -99% or 60% -90% diastole and 30% -60% systole by presetting phase intervals, e.g. 1% intervals. The existing method comprises the following steps: 1. and calculating the difference between the reconstructed image and the adjacent phase reconstructed image in the selected phase range, and taking the phase corresponding to the image with the smallest difference as the optimal phase. 2. And calculating SSIM (structural similarity) between the reconstructed image and the adjacent phase reconstructed image in the selected phase range, and taking the phase corresponding to the image with the maximum SSIM as the optimal phase. 3. And adopting the image definition of the region of interest, such as information entropy and the like, wherein the phase corresponding to the reconstructed image with the highest definition is the optimal phase.
The image difference calculation method and the structure similarity method are easy to be interfered by image noise. To reduce patient dose, cardiac scanning typically uses an ECG synchronized current modulation protocol, as shown in fig. 2. The doctor sets the interested phase according to the heart rate condition of the patient, exposes the heart rate by high current near the phase, exposes the heart rate by low current mA at other positions, and causes great noise difference in the reconstructed images due to dose difference. In order to ensure image noise uniformity, the time of the high current must be increased. The definition method needs to calculate aiming at a specific region of interest, inconsistent phenomenon exists among the positions and the forms of the regions of interest with different phases, and the algorithm for accurately extracting the regions of interest is complex and has large calculation amount.
Disclosure of Invention
The technical purpose is that: aiming at the defects of the existing method, the invention provides a method, equipment and a storage medium for automatically identifying the global optimal phase of cardiac CT imaging, which can quantitatively describe cardiac motion and has less influence of image noise on registration accuracy.
The technical scheme is as follows: in order to achieve the above purpose, the present invention provides the following technical solutions:
a global optimal phase automatic identification method for cardiac CT imaging comprises the following steps:
(1) Multiphase image reconstruction: selecting a plurality of phases at intervals in a cardiac cycle, and reconstructing according to CT scanning data corresponding to the selected phases to obtain reconstructed images corresponding to the phases; the set of all pixels in each reconstructed image is an image matrix, the rows of the image matrix correspond to the heights of the reconstructed images, the columns of the image matrix correspond to the widths of the reconstructed images, and the elements of the image matrix correspond to the pixels of the reconstructed images;
(2) Image registration: performing image registration on the reconstructed images of any two adjacent phases in the plurality of phases, and obtaining a deformation field of each pixel in the reconstructed images of any two adjacent phases in the plurality of phases after registration, wherein the deformation field obtained by registration is a deformation field in the heart motion X, Y and the Z direction;
(3) Motion estimation: according to the time interval between any two adjacent phases and the deformation field of each pixel in the corresponding image matrix, calculating to obtain the heart motion speed corresponding to each pixel in any one phase image matrix, and then calculating the heart motion speed corresponding to each pixel in all the phase image matrixes selected in the step (1);
obtaining a heart motion speed function by a fitting or interpolation method according to the heart motion speed corresponding to each pixel in the calculated multiple phase image matrixes, wherein the heart motion speed function is used for obtaining the heart motion speed corresponding to each pixel in the image matrix at any phase moment, and the heart motion speed is X, Y and the speed in the Z direction;
(4) And (3) optimal phase identification: calculating the heart motion speed corresponding to each pixel in the image matrix of any one of the phases selected in the step (1) according to the heart motion speed function, calculating the modes of heart motion speed vectors corresponding to all the pixels in the image matrix, and then calculating the average value of the modes, wherein the average value represents the average motion amplitude of the heart at any one of the phases, the modes of the heart motion speed vectors represent the motion amplitude of the heart motion speed at the selected phase, and the directions of the heart motion speed vectors represent the motion directions at the selected phase;
traversing all phases selected in the step (1), wherein the phase moment with the smallest mean value is the optimal phase moment.
Preferably, in the step (1), the plurality of phases are selected at equal intervals or unequal intervals, the image reconstruction interval is 0% -99% of the cardiac cycle, the phase selection range is 0-100, and the interval between each two phases is less than or equal to 2.
Preferably, in the step (2), the reconstructed images of any two adjacent phases are subjected to image registration according to the following formula:
sim＝α·MSE(img ref ,img reg )+β·MI(img ref ,img reg )+γ·||warp|| 2
wherein α, β and γ represent weights used to control different similarities, img ref To refer to the graph, img reg For the registration map of the floating map after transformation according to the deformation field, warp is the deformation field.
Preferably, in step (2), the method of image registration comprises 3D registration, decomposition into X-Y plane and X-Z plane registration or X-Y plane and Y-Z plane registration, artificial intelligence based registration.
Preferably, in step (2), the deformation fields resulting from the registration are the heart motion X, Y and the deformation fields in the Z direction.
Preferably, in step (3), the time interval t between two adjacent phases i ＝cycle*(phase i+1 -phase i ) Wherein i is the sequence number of the selected phase, cycle is the cardiac cycle, phase i For the selected phase, phase i+1 Is phase i Is a phase of the adjacent phase.
The invention provides an electronic device, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor realizes the global optimal phase automatic identification method for cardiac CT imaging according to any one of the preferred schemes when executing the program.
The invention also provides a computer readable storage medium, which is characterized in that computer executable instructions are stored, and the computer executable instructions are used for executing the automatic optimal phase identification method of cardiac CT according to any one of the preferred schemes.
The beneficial effects are that: compared with the prior art, the invention has the following advantages:
when multiple phases are selected, the invention can be selected at equal intervals or unequal intervals, the heart space motion speed is calculated according to the deformation field and the time interval between two adjacent phases, the amplitude and the direction of heart motion are given, the function about the heart motion speed can be obtained through fitting, and the heart can be quantitatively described.
Drawings
FIG. 1 is a schematic illustration of a real-time electrocardiogram-based cardiac scan acquisition data source;
FIG. 2 is a schematic diagram of an Electrocardiogram (ECG) synchronized current modulation protocol;
FIG. 3 is a flow chart of the technical scheme of the invention;
FIG. 4 is a schematic illustration of a registration method;
fig. 5 is a spatial motion velocity vector diagram of the heart at any instant.
Detailed Description
The present invention will now be described more fully hereinafter with reference to the accompanying drawings, in which preferred embodiments of the invention are shown, but in which the invention is not so limited.
As shown in fig. 3, the invention provides a global optimal phase automatic identification method for cardiac CT imaging, comprising the following steps:
1. multi-phase image reconstruction
Specifically, selecting a plurality of phases at intervals of one cardiac cycle, and reconstructing according to CT scanning data corresponding to the selected phases to obtain reconstructed images corresponding to the phases; the set of all pixels in the reconstructed image corresponding to the phases is an image matrix, the rows of the image matrix correspond to the height of the reconstructed image, the columns of the image matrix correspond to the width of the reconstructed image, and the elements of the image matrix correspond to the pixels of the reconstructed image.
The phase selection can be equal interval selection or unequal interval selection, the range of the phase selection is 0-100, the interval between each phase is less than or equal to 2, and the phase represents the position of the heart in the current cardiac cycle. The interval of image reconstruction is 0% -99% of the cardiac cycle, the electrocardiogram of the heart can be divided into cycles by R wave, namely the cardiac cycle, if the phase is about 45% of the current cardiac cycle, the heart is usually in the systolic phase, and if the phase is about 75% of the current cardiac cycle, the heart is usually in the diastolic phase.
Firstly, selecting a plurality of phases 2,4,6,8, 10, 12, 14, 16, 18, 20 and the like at equal intervals in 0-99% of a cardiac cycle. Reconstructing a plurality of reconstructed images img corresponding to the phases of 2,4,6,8, 10, 12, 14, 16, 18, 20 and the like according to the selected phases 1 、img 2 、img 3 、img 4 、img 5 、img 6 、img 7 、img 8 、img 9 、img 10 Etc.
2. Image registration
Specifically, image registration is performed on the reconstructed images of any two adjacent phases in the plurality of phases, a deformation field of each pixel in the reconstructed images of any two adjacent phases in the plurality of phases is obtained after registration, and the deformation fields obtained by registration are deformation fields in the heart motion X, Y and the Z direction. The task of the image registration technology is to determine the point-to-point mapping relation between two images obtained by the same scene under different shooting angles, time, depth or equipment, so that the position points of the same scene of the images are accurately corresponding, and the fusion of image information is realized.
There are various image similarity losses, commonly known as correlation coefficients (Correlation Coefficient, CC), normalized Correlation Coefficients (NCC), mutual information (Mutual Information, MI), mean Square Error (MSE), etc. When images have similar gray value distributions, MSE is typically used to evaluate the similarity of gray values, NCC and MI being more appropriate when gray values are transformed.
Cardiac scanning is a contrast enhanced scanning in which the concentration of contrast agent in the body changes over time, resulting in inconsistent image CT values, i.e. differences in gray values, between different phases. As shown in fig. 4, when the floating image and the reference image are aligned, deformation fields of all pixels in the floating image are calculated, then the floating image is transformed according to the deformation fields, whether the floating image and the reference image meet the similarity condition is judged, if yes, the deformation fields are output, and if not, the deformation fields of all pixels in the floating image are continuously updated and calculated based on the existing deformation fields.
The deformation field should be kept smooth, so the invention proposes the following similarity conditions to adapt to different gray value variations. The similarity criterion provided by the invention is that other two kinds of similarity are added on the basis of common MSE similarity, one kind of similarity is MI mutual information, the other kind of similarity is based on functions of deformation field vectors, the deformation field vectors are added, and the image registration can be realized on the premise that the whole deformation field is as small as possible.
sim＝α·MSE(img ref ,img reg )+β·MI(img ref ,img reg )+γ·||warp|| 2
Wherein α, β and γ represent weights used to control different similarities, img reg Img as reference image reg For the registered image of the floating image transformed from the deformation field, warp is the deformation field.
MI(img ref ,img reg )＝H(img ref )+H(img reg )-H(img ref ,img reg )
Wherein H (img) ref ) And H (img) reg ) Respectively representing the image img ref And img reg Information entropy of (1), H (img) reg ,img reg ) Is the joint entropy of the image.
p i Is the pixel point with the value i is img ref Probability of occurrence, p j Is the pixel point with the value j is img reg Probability of occurrence.Is the pixel point at the same position, img ref The value is i, img reg Probability at value j. The probability may be obtained by histogram statistics.
The positions of two adjacent images are equal during registration, the two adjacent images are respectively used as a floating image and a reference image, and the purpose of registration is achieved through multiple iterations. The image registration method comprises 3D registration, X-Y plane registration and X-Z plane registration or X-Y plane registration and Y-Z plane registration, and artificial intelligence registration.
The image registration method which is divided into X-Y plane and X-Z plane registration or X-Y plane and Y-Z plane registration and is mainly based on characteristics, and the basic flow can be summarized into the following four steps: feature detection, feature matching, conversion model estimation and image fusion. The image registration technology is widely applied, and representative feature extraction algorithms mainly comprise SIFT, SURF, kaze, BRISK algorithm and the like.
The 3D registration flow can be decomposed into two 2D registration processes, so that the calculated amount is reduced, and the calculation speed is improved.
Based on artificial intelligence registration: network models such as RegNet, unet, voxelmorph and the like can be constructed, and the models are trained and deployed by using the above loss functions.
For example phase 2 and phase 4 corresponding reconstructed images img 1 And img 2 Image registration is carried out to obtain deformation field of each pixel as warp_x 1 (x,y,z)、wrp_y 1 (x, y, z) and warp_z 1 (x, y, z), phase 4 and phase 6 corresponding reconstructed images img 2 And img 3 Image registration is carried out to obtain deformation field of each pixel as warp_x 2 (x,y,z)、warp_y 2 (x, y, z) and warp_z 2 And (x, y, z) and the like, the deformation field of each pixel after image registration of any two adjacent phases of the selected multiple phases can be obtained, wherein x, y and z are 3D space pixel indexes of the image. Because the motion of the heart is a three-dimensional motion, the use of a three-dimensional deformation field can better approximate the actual motion.
The traditional image registration method has no good effect when registering if certain data and models are not matched, and is usually solved by adopting a mode of modifying the difference measurement, but the image registration based on artificial intelligence is automatically determined by adopting a machine learning mode rather than adopting a mode of manually modifying the models, namely, the difference measurement standard is not explicitly defined.
3. Motion estimation
Specifically, according to the time interval between any two adjacent phases and the deformation field of each pixel in the corresponding image matrix, calculating to obtain the heart motion speed corresponding to each pixel in the image matrix of any one phase, then calculating the heart motion speed corresponding to each pixel in all the phase image matrices selected in the step (1), obtaining a function about the heart motion speed according to the heart motion speed corresponding to each pixel in the calculated phase image matrices by a fitting or interpolation method, and obtaining the heart motion speed corresponding to each pixel in the image matrix at any phase moment according to the curve function, wherein the heart motion speed is X, Y and the Z-direction speed.
For example, the time interval for calculating phases 2 and 4 is t 1 ，t 1 =cycle (4-2), where eye is the cardiac cycle, the time interval t of phases 2 and 4 assuming a cycle of 0.75 heart beats 1 =1.5, then the heart motion velocity corresponding to each pixel of the heart at phase 2 time is v_x 1 (x,y,z)＝warp_x 1 (x,y,z)/t 1 、v_y 1 (x,y,z)＝warp_y 1 (x,y,z)/t 1 And v_z 1 (x,y,z)＝warp_z 1 (x,y,z)/t 1 Where x, y, z are 3D spatial pixel indices of the image. The heart motion speed corresponding to each pixel at the time of the phases of 4,6 and 8 can be obtained by the same method, and the heart motion speed corresponding to each pixel at any time can be obtained by fitting or interpolating according to the calculated heart motion speeds of the phases.
4. Optimal phase identification
Specifically, traversing the heart motion speed corresponding to each pixel in the image matrix of any one of the selected phases, calculating the modes of heart motion speed vectors corresponding to all the pixels in the image matrix, and then calculating the mean value of the modes, wherein the mean value of the heart motion speed vectors represents the average motion amplitude of the heart at any one of the selected phases, the modes of the heart motion speed vectors represent the motion amplitude of the heart at the selected phase, and the directions of the heart motion speed vectors represent the motion directions at the selected phase; traversing all phases selected in the step (1), wherein the phase moment with the smallest mean value is the optimal phase moment.
Fig. 5 is a schematic diagram of a spatial motion velocity of a heart, where OP vectors represent motion velocities of the heart corresponding to each pixel at any one phase time, vectors v_x, v_y, v_z represent components of the motion velocity of the heart in X, Y and Z directions, OQ is a modulus of the OP vectors, represents a motion amplitude of the heart at the phase time, and a direction of the OP vectors represents a motion direction of the heart at the phase time. Traversing the heart motion speed of each pixel in the image matrix at the phase moment, calculating OQ according to the components v_x, v_y and v_z of the OP vector in X, Y and Z directions, and then calculating the average value of all the OQ as the average motion amplitude of the heart at the phase moment; traversing all phases selected in the reconstruction interval, wherein the time with the minimum average value of the OQ is the optimal phase time.
When reconstructing the multiphase image, the invention scans and acquires projection data of the whole pericardial area, reconstructs the image of the whole pericardial area, and performs image registration on the image of the whole pericardial area, namely the global area refers to the image registration range. The registration is aimed at the whole heart, and does not need to be aimed at blood vessels alone, so that the step of segmenting blood vessels in the prior art is omitted, and the reconstructed image is segmented roughly according to the pericardial area before the image registration. Because the volume of the pericardium is large, the requirement on the segmentation precision is not high, and the deformation fields of the heart motion X, Y and the Z direction are obtained through image registration, the heart motion is quantitatively described, the influence and the interference of the image noise on the registration precision are small, the region of interest does not need to be set in advance, and the calculation amount is small.
The invention provides an electronic device, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor realizes any one of the above global optimal phase automatic identification method for cardiac CT imaging when executing the program. The memory may be various types of memory, such as random access memory, read only memory, flash memory, etc. The processor may be various types of processors, such as a central processing unit, a microprocessor, a digital signal processor, or an image processor, etc.
The invention also provides a computer readable storage medium storing computer executable instructions for executing any one of the above-mentioned methods for automatically identifying global optimal phases of cardiac CT imaging. The storage medium includes: a usb disk, a removable hard disk, a ROM, a RAM, a magnetic disk, or an optical disk, etc.
The foregoing is only a preferred embodiment of the invention, it being noted that: it will be apparent to those skilled in the art that various modifications and adaptations can be made without departing from the principles of the present invention, and such modifications and adaptations are intended to be comprehended within the scope of the invention.
Claims (7)
1. The global optimal phase automatic identification method for cardiac CT imaging is characterized by comprising the following steps:
(1) Multiphase image reconstruction: selecting a plurality of phases at intervals in a cardiac cycle, and reconstructing according to CT scanning data corresponding to the selected phases to obtain reconstructed images corresponding to the phases; the set of all pixels in each reconstructed image is an image matrix, the rows of the image matrix correspond to the heights of the reconstructed images, the columns of the image matrix correspond to the widths of the reconstructed images, and the elements of the image matrix correspond to the pixels of the reconstructed images;
(2) Image registration: performing image registration on the reconstructed images of any two adjacent phases in the plurality of phases, and obtaining a deformation field of each pixel in the reconstructed images of any two adjacent phases in the plurality of phases after registration, wherein the deformation field obtained by registration is a deformation field in the heart motion X, Y and the Z direction;
(3) Motion estimation: according to the time interval between any two adjacent phases and the deformation field of each pixel in the corresponding image matrix, calculating to obtain the heart motion speed corresponding to each pixel in any one phase image matrix, and then calculating the heart motion speed corresponding to each pixel in all the phase image matrixes selected in the step (1);
obtaining a heart motion speed function by a fitting or interpolation method according to the heart motion speed corresponding to each pixel in the calculated multiple phase image matrixes, wherein the heart motion speed function is used for obtaining the heart motion speed corresponding to each pixel in the image matrix at any phase moment, and the heart motion speed is X, Y and the speed in the Z direction;
(4) And (3) optimal phase identification: calculating the heart motion speed corresponding to each pixel in the image matrix of any one of the phases selected in the step (1) according to the heart motion speed function, calculating the modes of heart motion speed vectors corresponding to all the pixels in the image matrix, and then calculating the average value of the modes, wherein the average value represents the average motion amplitude of the heart at any one of the phases, the modes of the heart motion speed vectors represent the motion amplitude of the heart motion speed at the selected phase, and the directions of the heart motion speed vectors represent the motion directions at the selected phase;
traversing all phases selected in the step (1), wherein the phase moment with the smallest mean value is the optimal phase moment.
2. The method for automatically identifying global optimal phases of cardiac CT imaging according to claim 1, wherein in the step (1), a plurality of phases are selected at equal intervals or unequal intervals, an image reconstruction interval is 0% -99% of a cardiac cycle, a phase selection range is 0-100, and an interval between each two phases is less than or equal to 2.
3. The method for automatically identifying global optimal phases for cardiac CT imaging according to claim 1, wherein in the step (2), the reconstructed images of any two adjacent phases are subjected to image registration according to the following formula:
sim＝α·MSE(img ref ,img reg )+β·MI(img ref ,img reg )+γ·||warp|| 2
wherein α, β and γ represent weights used to control different similarities, img ref To refer to the graph, img reg For the registration map of the floating map after transformation according to the deformation field, warp is the deformation field.
4. A global optimal phase automatic identification method for cardiac CT imaging according to claim 3, wherein in step (2), the method of image registration comprises 3D registration, decomposition into X-Y plane and X-Z plane registration or X-Y plane and Y-Z plane registration, artificial intelligence based registration.
5. The method for automatically identifying global optimal phases for cardiac CT imaging according to claim 1, wherein in step (3), a time interval t between two adjacent phases i ＝cycle*(phase i+1 -phase i ) Wherein i is the sequence number of the selected phase, cycle is the cardiac cycle, phase i For the selected phase, phase i+1 Is phase i Is a phase of the adjacent phase.
6. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing a global optimal phase auto-identification method for cardiac CT imaging as claimed in any one of claims 1-5 when the program is executed.
7. A computer readable storage medium having stored thereon computer executable instructions for performing a cardiac CT imaging global optimum phase automatic identification method according to any one of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311009213.6A CN116849691B (en) | 2023-08-11 | 2023-08-11 | Method, equipment and storage medium for automatically identifying global optimal phase of cardiac CT imaging |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311009213.6A CN116849691B (en) | 2023-08-11 | 2023-08-11 | Method, equipment and storage medium for automatically identifying global optimal phase of cardiac CT imaging |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116849691A true CN116849691A (en) | 2023-10-10 |
| CN116849691B CN116849691B (en) | 2024-03-12 |
Family
ID=88232373
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311009213.6A Active CN116849691B (en) | 2023-08-11 | 2023-08-11 | Method, equipment and storage medium for automatically identifying global optimal phase of cardiac CT imaging |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116849691B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117115577A (en) * | 2023-10-23 | 2023-11-24 | 南京安科医疗科技有限公司 | Cardiac CT projection domain optimal phase identification method, equipment and medium |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102007029731A1 (en) * | 2007-06-27 | 2009-01-02 | Siemens Ag | Automatic determination of an optimum heart phase for computer tomography reconstruction, masks a motion map from image data sets to determine movement minima for the systolic and diastolic end phases |
| CN106920234A (en) * | 2017-02-27 | 2017-07-04 | 北京连心医疗科技有限公司 | A kind of method of the automatic radiotherapy planning of combined type |
| CN109345606A (en) * | 2018-09-27 | 2019-02-15 | 上海联影医疗科技有限公司 | Cardiac image method for reconstructing, device, computer equipment and readable storage medium storing program for executing |
| CN109389653A (en) * | 2018-09-27 | 2019-02-26 | 上海联影医疗科技有限公司 | Cardiac image method for reconstructing, device, computer equipment and readable storage medium storing program for executing |
| CN109754448A (en) * | 2018-12-29 | 2019-05-14 | 深圳安科高技术股份有限公司 | A kind of CT heart scanning artifact correction method and its system |
| CN110148150A (en) * | 2019-06-20 | 2019-08-20 | 深圳大学 | It is dense to connect asymmetric hierarchical network training method and heart movement field estimation method |
| CN111161333A (en) * | 2019-12-12 | 2020-05-15 | 中国科学院深圳先进技术研究院 | Liver respiratory motion model prediction method and device and storage medium |
| CN112686935A (en) * | 2021-01-12 | 2021-04-20 | 武汉大学 | Airborne depth sounding radar and multispectral satellite image registration method based on feature fusion |
| CN116012344A (en) * | 2023-01-29 | 2023-04-25 | 东北林业大学 | Cardiac magnetic resonance image registration method based on mask self-encoder CNN-transducer |
-
2023
- 2023-08-11 CN CN202311009213.6A patent/CN116849691B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102007029731A1 (en) * | 2007-06-27 | 2009-01-02 | Siemens Ag | Automatic determination of an optimum heart phase for computer tomography reconstruction, masks a motion map from image data sets to determine movement minima for the systolic and diastolic end phases |
| CN106920234A (en) * | 2017-02-27 | 2017-07-04 | 北京连心医疗科技有限公司 | A kind of method of the automatic radiotherapy planning of combined type |
| CN109345606A (en) * | 2018-09-27 | 2019-02-15 | 上海联影医疗科技有限公司 | Cardiac image method for reconstructing, device, computer equipment and readable storage medium storing program for executing |
| CN109389653A (en) * | 2018-09-27 | 2019-02-26 | 上海联影医疗科技有限公司 | Cardiac image method for reconstructing, device, computer equipment and readable storage medium storing program for executing |
| CN109754448A (en) * | 2018-12-29 | 2019-05-14 | 深圳安科高技术股份有限公司 | A kind of CT heart scanning artifact correction method and its system |
| CN110148150A (en) * | 2019-06-20 | 2019-08-20 | 深圳大学 | It is dense to connect asymmetric hierarchical network training method and heart movement field estimation method |
| CN111161333A (en) * | 2019-12-12 | 2020-05-15 | 中国科学院深圳先进技术研究院 | Liver respiratory motion model prediction method and device and storage medium |
| CN112686935A (en) * | 2021-01-12 | 2021-04-20 | 武汉大学 | Airborne depth sounding radar and multispectral satellite image registration method based on feature fusion |
| CN116012344A (en) * | 2023-01-29 | 2023-04-25 | 东北林业大学 | Cardiac magnetic resonance image registration method based on mask self-encoder CNN-transducer |
Non-Patent Citations (1)
| Title |
|---|
| HIDEHARU MIURA 等: "Impact of deformable image registration accuracy on thoracic images with different regularization weight parameter settings", PHYSICA MEDICA, vol. 42, pages 108 - 111, XP085286237, DOI: 10.1016/j.ejmp.2017.09.122 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117115577A (en) * | 2023-10-23 | 2023-11-24 | 南京安科医疗科技有限公司 | Cardiac CT projection domain optimal phase identification method, equipment and medium |
| CN117115577B (en) * | 2023-10-23 | 2023-12-26 | 南京安科医疗科技有限公司 | Cardiac CT projection domain optimal phase identification method, equipment and medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116849691B (en) | 2024-03-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Armanious et al. | Unsupervised medical image translation using cycle-MedGAN | |
| Painchaud et al. | Cardiac segmentation with strong anatomical guarantees | |
| Xia et al. | Super-resolution of cardiac MR cine imaging using conditional GANs and unsupervised transfer learning | |
| CN109754448B (en) | CT cardiac scanning artifact correction method and system | |
| US20070276245A1 (en) | System And Method For Automated Boundary Detection Of Body Structures | |
| CN116849691B (en) | Method, equipment and storage medium for automatically identifying global optimal phase of cardiac CT imaging | |
| CN111127504A (en) | Heart medical image segmentation method and system for atrial septal occlusion patient | |
| CN116740218B (en) | Heart CT imaging image quality optimization method, device and medium | |
| WO2022086910A1 (en) | Anatomically-informed deep learning on contrast-enhanced cardiac mri | |
| CN114037626A (en) | Blood vessel imaging method, device, equipment and storage medium | |
| CN112070752A (en) | Method, device and storage medium for segmenting auricle of medical image | |
| CN115861132B (en) | Blood vessel image correction method, device, medium and equipment | |
| CN106910182B (en) | Segmentation method of blood pool in end diastole image in cardiac function magnetic resonance image | |
| Nayak et al. | Morphology based iterative back-projection for super-resolution reconstruction of image | |
| US20210150778A1 (en) | Motion compensated cardiac valve reconstruction | |
| CN115861172A (en) | Wall motion estimation method and device based on self-adaptive regularized optical flow model | |
| US20210287375A1 (en) | System architecture and method of processing images | |
| Shao et al. | Semantic segmentation method of 3D liver image based on contextual attention model | |
| Choy et al. | Extracting endocardial borders from sequential echocardiographic images | |
| Guo et al. | Left ventricle volume measuring using echocardiography sequences | |
| Romero-Pacheco et al. | Estimating Echocardiographic Myocardial Strain of Left Ventricle with Deep Learning | |
| Wang et al. | 3D cardiac motion reconstruction from CT data and tagged MRI | |
| CN113112507B (en) | Perfusion image analysis method, system, electronic equipment and storage medium | |
| US20240153089A1 (en) | Systems and methods for processing real-time cardiac mri images | |
| Hou et al. | Segmentation of Multi-Center Cardiac Magnetic Resonance Images Based on Improved DeepLabV3+ |
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 |