CN114612533A - Head multi-mode CT image registration method and device and radiotherapy equipment - Google Patents
Head multi-mode CT image registration method and device and radiotherapy equipment Download PDFInfo
- Publication number
- CN114612533A CN114612533A CN202210242421.XA CN202210242421A CN114612533A CN 114612533 A CN114612533 A CN 114612533A CN 202210242421 A CN202210242421 A CN 202210242421A CN 114612533 A CN114612533 A CN 114612533A
- Authority
- CN
- China
- Prior art keywords
- head
- image
- curve
- images
- point
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 42
- 238000001959 radiotherapy Methods 0.000 title claims abstract description 15
- 210000003128 head Anatomy 0.000 claims abstract description 147
- 210000001214 frontal sinus Anatomy 0.000 claims abstract description 44
- 230000009467 reduction Effects 0.000 claims abstract description 21
- 238000009499 grossing Methods 0.000 claims abstract description 14
- 238000007781 pre-processing Methods 0.000 claims abstract description 12
- 230000001815 facial effect Effects 0.000 claims abstract description 8
- 238000004445 quantitative analysis Methods 0.000 claims abstract description 7
- 238000010606 normalization Methods 0.000 claims description 17
- 238000000605 extraction Methods 0.000 claims description 14
- 238000001914 filtration Methods 0.000 claims description 10
- 238000000354 decomposition reaction Methods 0.000 claims description 6
- 230000001965 increasing effect Effects 0.000 claims description 5
- 230000008569 process Effects 0.000 claims description 3
- 210000003625 skull Anatomy 0.000 abstract description 4
- 238000005516 engineering process Methods 0.000 abstract description 3
- 238000009795 derivation Methods 0.000 abstract 1
- 238000012545 processing Methods 0.000 description 9
- 206010028980 Neoplasm Diseases 0.000 description 8
- 238000004458 analytical method Methods 0.000 description 4
- 238000003384 imaging method Methods 0.000 description 4
- 210000000988 bone and bone Anatomy 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 230000002708 enhancing effect Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 210000002454 frontal bone Anatomy 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000000056 organ Anatomy 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Images
Classifications
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/10—Image enhancement or restoration by non-spatial domain filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/20—Image enhancement or restoration by the use of local operators
-
- G06T5/70—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—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
- G06T2207/10072—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
- G06T2207/20048—Transform domain processing
- G06T2207/20064—Wavelet transform [DWT]
-
- 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
- G06T2207/30016—Brain
Abstract
The invention discloses a head multi-modal CT image registration method, a head multi-modal CT image registration device and radiotherapy equipment, wherein the head multi-modal CT image registration method specifically comprises the following steps: the method comprises the steps of obtaining a reference image of the head and an image to be registered of the head, respectively extracting facial edges after preprocessing the image to obtain a head contour curve, then carrying out quantitative analysis on the head contour curve to obtain a head characteristic curve, conducting noise reduction and smoothing, then conducting derivation on each point of two smooth head characteristic curves, taking a point with a first derivative being zero as the frontal part of the surface of the head, taking the point as the center, respectively adding a specified number of pixels in the upper, lower, left and right directions, extracting frontal sinus regions of two images, and registering the two images by taking the mass center coordinates of the frontal sinus regions of the two images as the basis. The invention uses the biological characteristics of the human skull as the basis of image registration by using the image recognition technology, and corrects the spatial position by using the position information of the characteristic points, thereby reducing the positioning error and realizing accurate treatment.
Description
Technical Field
The invention belongs to the field of tumor radiotherapy based on image guidance, and particularly relates to a head multi-mode CT image registration method and device and radiotherapy equipment.
Background
Precision radiation therapy is one of the important methods in the current treatment of tumors. In the course of radiotherapy, the positioning error existing in each treatment will cause the change of the target area (tumor area), and the inaccuracy of the target area not only can cause the target area to miss the irradiation, but also can cause the high-energy ray to move out of the target area and even move into the dangerous organ area, thereby causing serious complications or sequelae. ICRU (International Commission on Radiation Units and measurements) report No. 24, which states: deviation of the target irradiation dose by 5% may lead to uncontrolled primary focus or increased complications. How to efficiently and accurately align the tumor region of a patient to a beam current is one of the difficulties in radiotherapy.
Currently, the manual registration is still mainly used clinically, and two images of different modalities, namely a CT reconstructed image (also called as a reference image or a DRR image) and an onboard CT real-time imaging (also called as a registration image, and the ray energy is KV level or MV level imaging) which are used as the basis for judging the tumor position by a doctor are manually registered. The manual registration is time-consuming and labor-consuming, the registration accuracy mainly depends on the personal experience of a doctor, and the registration consistency is poor due to the major influence. Some researches for identifying the characteristic region by adopting a deep learning method to perform automatic registration also exist, but the deep learning method has identification difference on imaging in different modes, has higher requirement on the number of samples, and has the problems of poorer identification precision, lower accuracy and the like.
Disclosure of Invention
In order to solve the technical problem, the invention provides a head multi-mode CT image registration method, a head multi-mode CT image registration device and radiotherapy equipment.
In order to achieve the purpose, the technical scheme of the invention is as follows:
on one hand, the invention discloses a head multi-mode CT image registration method, which specifically comprises the following steps:
s1: acquiring a reference image of the head and an image to be registered of the head, and preprocessing the two acquired images before feature extraction;
s2: respectively extracting facial edges of the two preprocessed images to obtain corresponding head contour curves;
s3: carrying out quantitative analysis on the two head contour curves respectively to obtain corresponding head characteristic curves;
s4: carrying out noise reduction smoothing treatment on the two head characteristic curves;
s5: deriving each point of the two smooth head characteristic curves, wherein the point with the first derivative being zero is the frontal part of the head surface, and taking the point as the center, increasing the pixels with the specified number respectively from top to bottom, left to right, and extracting frontal sinus regions of the two images;
s6: and calculating the mass center coordinates of the frontal sinus region of the reference image and the image to be registered, and registering the two images by taking the mass center coordinates of the frontal sinus region of the two images as the basis.
On the basis of the technical scheme, the following improvements can be made:
preferably, the pretreatment in S1 includes: one or more of cropping, size normalization, histogram normalization.
Preferably, the establishment of the head characteristic curve in S3 specifically includes the following steps:
and establishing a coordinate system, wherein the vertex of the head in the head contour curve is used as an origin, the longitudinal distance of each point on the head contour curve is used as the numerical value of the x axis or the y axis, and the transverse distance of each point on the contour curve is used as the numerical value of the y axis or the x axis.
Preferably, the noise reduction smoothing process in S4 specifically includes the following steps:
and performing wavelet decomposition on the extracted head characteristic curve, analyzing the characteristic region frequency characteristics of the head characteristic curve, and filtering curve noise in a frequency domain to ensure that the curve is smooth.
In another aspect, the present invention further discloses a head multi-modality CT image registration apparatus, including:
the image acquisition module is used for acquiring a reference image of the head and an image to be registered of the head, and preprocessing the two acquired images before feature extraction;
the contour curve extraction module is used for respectively extracting the facial edges of the two preprocessed images to obtain corresponding head contour curves;
the characteristic curve acquisition module is used for respectively carrying out quantitative analysis on the two head contour curves to obtain corresponding head characteristic curves;
the noise reduction module is used for carrying out noise reduction smoothing treatment on the two head characteristic curves;
the frontal sinus region extraction module is used for deriving each point of the two smooth head characteristic curves, the point with the first derivative being zero is the frontal part of the head surface, the point is taken as the center, the pixels with the specified number are respectively added in the upper, lower, left and right directions, and the frontal sinus regions of the two images are extracted;
and the registration module is used for calculating the mass center coordinates of the frontal sinus region of the reference image and the image to be registered respectively, and registering the two images by taking the mass center coordinates of the frontal sinus region of the two images as the basis.
Preferably, the image acquisition module can perform one or more preprocessing operations of cropping, size normalization and histogram normalization on the acquired reference image and the image to be registered.
Preferably, the characteristic curve obtaining module establishes the head characteristic curve by the following steps:
and establishing a coordinate system, wherein the vertex of the head in the head contour curve is used as an origin, the longitudinal distance of each point on the head contour curve is used as the numerical value of the x axis or the y axis, and the transverse distance of each point on the contour curve is used as the numerical value of the y axis or the x axis.
Preferably, the noise reduction module performs noise reduction smoothing by:
and performing wavelet decomposition on the extracted head characteristic curve, analyzing the characteristic region frequency characteristics of the head characteristic curve, and filtering curve noise in a frequency domain to ensure that the curve is smooth.
In addition, the invention also discloses a radiotherapy device, which carries out registration by using any one of the multi-modal head CT image registration methods, or comprises the following steps: any one of the above multi-modality head CT image registration devices.
The invention relates to a method and a device for registering head multi-mode CT images and radiotherapy equipment, which use an image recognition technology to use the biological characteristics of the skull of a human body as the basis of image registration and use the position information of characteristic points to correct the spatial positions of a reference image and an image to be registered, thereby effectively reducing the positioning error and leading the tumor part of a patient to be treated more accurately.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.
Fig. 1 is a flowchart of a head multi-modality CT image registration method according to an embodiment of the present invention.
Fig. 2 is a reference image and an image to be registered after histogram normalization according to an embodiment of the present invention (where (a) is the reference image and (b) is the image to be registered);
FIG. 3 is a head contour curve of an image to be registered according to an embodiment of the present invention;
fig. 4 is a head characteristic curve of the image to be registered according to the embodiment of the present invention after coordinate transformation;
FIG. 5 shows the high-frequency and low-frequency component analysis of DB5 wavelet on three scales (wherein (a) is low-frequency component analysis, a1 is scale one low frequency, a2 is scale two low frequency, a3 is scale three low frequency, (b) is high-frequency component analysis, d1 is scale one high frequency, d2 is scale two high frequency, and d3 is scale three high frequency);
fig. 6 is a head characteristic curve of the noise-reduced and smoothed image to be registered according to the embodiment of the present invention;
FIG. 7 is an extracted frontal sinus region provided by an embodiment of the present invention (where (a) is a reference image and (b) is an image to be registered);
FIG. 8 shows a frontal sinus region after wiener filtering (where (a) is a reference image and (b) is an image to be registered) according to an embodiment of the present invention;
fig. 9 is a frontal sinus region after binarization processing according to an embodiment of the present invention (where (a) is a reference image and (b) is an image to be registered);
FIG. 10 is a frontal sinus region after the opening operation and the extraction of the connected region to the margin provided by the embodiment of the invention (wherein, (a) is a reference image and, (b) is an image to be registered);
fig. 11 is a display of the extracted centroid coordinates in the complete image according to the embodiment of the present invention (where (a) is a reference image and (b) is an image to be registered).
Detailed Description
Preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings.
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The use of the ordinal terms "first," "second," "third," etc., to describe a common object merely indicate that different instances of like objects are being referred to, and are not intended to imply that the objects so described must be in a given sequence, either temporally, spatially, in ranking, or in any other manner.
Also, the expression "comprising" an element is an expression of "open" which merely means that there is a corresponding component, and should not be interpreted as excluding additional components.
In order to achieve the object of the present invention, in some embodiments of a head multi-modality CT image registration method, a head multi-modality CT image registration device, and a radiotherapy apparatus, the head multi-modality CT image registration method specifically includes the following steps:
s1: acquiring a reference image of the head and an image to be registered of the head, and preprocessing the two acquired images before feature extraction;
s2: respectively extracting facial edges of the two preprocessed images to obtain corresponding head contour curves;
s3: carrying out quantitative analysis on the two head contour curves respectively to obtain corresponding head characteristic curves;
s4: carrying out noise reduction smoothing treatment on the two head characteristic curves;
s5: deriving each point of the two smooth head characteristic curves, wherein the point with the first derivative being zero is the frontal part of the head surface, and taking the point as the center, increasing the pixels with the specified number respectively from top to bottom, left to right, and extracting frontal sinus regions of the two images;
s6: and calculating the mass center coordinates of the respective frontal sinus regions of the reference image and the image to be registered, and registering the two images by taking the mass center coordinates of the frontal sinus regions of the two images as a basis.
The multi-mode CT image registration method in the prior art has the defects of low efficiency, low precision and the like. For the whole sagittal plane head CT image, a frontal sinus region is extracted, and an algorithm with strong adaptability is difficult to be provided due to the characteristic of ray imaging.
The inventor finds that the frontal sinus is a sinus cavity between the inner and outer bone plates behind the brow arch of the frontal bone and is a cavity in the bone in nature, so that the frontal sinus is fixed in position, and the internal cavity cannot attenuate rays, so that the frontal sinus presents uniform structural characteristics in radiographic images of different modalities and does not present different structural characteristics in different modalities like the skin of the body surface.
In S5, since the frontal sinus region corresponds to the frontal portion of the head surface and appears as the first vertex on the head characteristic curve, the point where the first derivative is zero corresponds to the frontal portion of the head surface, and the frontal sinus region can be extracted by adding a certain number of pixels to the point, up and down, left and right, respectively, as the center. The invention uses frontal sinus position in human skull as characteristic region as the basis of registration.
The head multi-mode CT image registration method corrects the spatial positions of the reference image and the image to be registered by utilizing the extracted information of the position, the size and the like of the frontal sinus region, thereby achieving the purpose of reducing the positioning error and leading the tumor part of the patient to be treated more accurately.
Further, on the basis of the above embodiment of the method, the preprocessing in S1 includes: one or more of cropping, size normalization, and histogram normalization. Cropping can effectively remove non-relevant areas, and histogram normalization is performed on the reference image and the image to be registered, so that the reference image and the image to be registered have the same gray distribution and are prepared for subsequent registration.
Further, on the basis of the above embodiment of the method, S2 may perform facial edge extraction by using a method of calculating a gradient, so as to obtain a corresponding head contour curve.
Further, on the basis of the above method embodiment, the establishment of the head characteristic curve in S3 specifically includes the following steps:
and establishing a coordinate system, wherein the vertex of the head in the head contour curve is used as an origin, the longitudinal distance of each point on the head contour curve is used as the numerical value of the x axis or the y axis, and the transverse distance of each point on the contour curve is used as the numerical value of the y axis or the x axis.
Further, on the basis of the above embodiment of the method, the noise reduction smoothing process in S4 specifically includes the following steps:
and performing wavelet decomposition on the extracted head characteristic curve, analyzing the characteristic region frequency characteristics of the head characteristic curve, and filtering curve noise in a frequency domain to ensure that the curve is smooth.
Further, on the basis of the above embodiment of the method, the following steps are further included between S5 and S6:
cutting an original reference image and an original image to be registered to obtain a subgraph only containing a frontal sinus region;
carrying out wiener filtering on the two sub-images to enhance the frontal sinus region in the sub-image;
further carrying out binarization processing on the two sub-images;
and further carrying out edge processing on the two sub-graphs, and eliminating the influence of the area connected with the edge by adopting a method of setting the communication area connected with the edge to be zero.
The embodiment of the invention also discloses a head multi-mode CT image registration device, which comprises:
the image acquisition module is used for acquiring a reference image of the head and an image to be registered of the head, and preprocessing the two acquired images before feature extraction;
the contour curve extraction module is used for respectively extracting the facial edges of the two preprocessed images to obtain corresponding head contour curves;
the characteristic curve acquisition module is used for respectively carrying out quantitative analysis on the two head contour curves to obtain corresponding head characteristic curves;
the noise reduction module is used for carrying out noise reduction smoothing treatment on the two head characteristic curves;
the frontal sinus region extraction module is used for deriving each point of the two smooth head characteristic curves, the point with the first derivative being zero is the frontal part of the head surface, the point is taken as the center, the pixels with the specified number are respectively added in the upper, lower, left and right directions, and the frontal sinus regions of the two images are extracted;
and the registration module is used for calculating the mass center coordinates of the frontal sinus region of the reference image and the image to be registered respectively, and registering the two images by taking the mass center coordinates of the frontal sinus region of the two images as the basis.
Further, on the basis of the above device embodiment, the image acquisition module can perform one or more preprocessing operations of cropping, size normalization and histogram normalization on the acquired reference image and the image to be registered.
Further, on the basis of the above device embodiment, the characteristic curve obtaining module establishes the head characteristic curve by the following steps:
and establishing a coordinate system, wherein the vertex of the head in the head contour curve is used as an origin, the longitudinal distance of each point on the head contour curve is used as the numerical value of the x axis or the y axis, and the transverse distance of each point on the contour curve is used as the numerical value of the y axis or the x axis.
Further, on the basis of the above device embodiment, the noise reduction module performs noise reduction smoothing processing by:
and performing wavelet decomposition on the extracted head characteristic curve, analyzing the characteristic region frequency characteristics of the head characteristic curve, and filtering curve noise in a frequency domain to ensure that the curve is smooth.
Further, on the basis of the above embodiment of the apparatus, the head multi-modality CT image registration apparatus further includes the following modules:
the cutting module is used for cutting the original reference image and the original image to be registered to obtain a subgraph only containing a frontal sinus region;
the enhancing module is used for carrying out wiener filtering on the two sub-images and enhancing the frontal sinus region in the sub-images;
the binarization processing module is used for carrying out binarization processing on the two sub-images;
and the edge processing module is used for carrying out edge processing on the two sub-graphs and eliminating the influence of the area connected with the edge by adopting a method of setting the communication area connected with the edge to be zero.
The embodiment of the invention also discloses radiotherapy equipment, which carries out registration by using the head multi-mode CT image registration method disclosed by any embodiment, or comprises the following steps: the head multi-modality CT image registration device disclosed by any one of the above embodiments.
A specific embodiment is described below, which uses the DRR image as the reference image and the KV image as the image to be registered, and it should be noted that in other embodiments, the MV image may also be used as the image to be registered.
The registration method is as follows:
1) obtaining a DRR image and a KV image;
2) histogram normalization processing is carried out on the DRR image and the KV image, and the normalized result, as shown in FIG. 2, can effectively eliminate the difference of the gray level distribution of the images in different modes through the histogram normalization;
3) background regions of the image containing no information are nearly white, and the gray scale value in the 16-bit CT image is a numerical value close to 65535. Therefore, the head and the background have a relatively clear boundary, and extracting the boundary is a head contour curve, because the gray values of points near the boundary of the feature region have large differences, the mathematical expression is that the gradient value at the point is large, and the contour edge of the head is determined by a method for calculating the gradient, taking a3 × 3 sub-region in the image as an example, the formula is as follows:
G(x,y)=|fx′|+|fy′|;
g (x, y) is the gradient at the central point;
fx' is the partial derivative in the x-direction;
fy' is the partial derivative in the y-direction;
since the image is a two-bit discrete function, the following steps are performed:
the extracted head contour curve is shown in fig. 3;
4) establishing a coordinate system, taking the vertex of the head in the head contour curve as an origin, taking the longitudinal distance of each point on the head contour curve as a numerical value of an x axis, and taking the transverse distance of each point on the contour curve as a numerical value of a y axis to obtain a corresponding head characteristic curve, as shown in fig. 4;
5) the head characteristic curves contain more noise, DB5 wavelets are used for analyzing the two head characteristic curves in a frequency domain, and the frequency characteristics of characteristic areas of the two head characteristic curves are analyzed;
in the formula: f (t) is a head characteristic curve;
{Cf(m,n)}m,n∈Zthe coefficient is wavelet transform (comprising two parts of a high-pass filter and a low-pass filter);
a is a scale factor;
b is a translation factor;
high frequency, low frequency coefficients on three scales are shown in fig. 5, and the present invention denoises the frequency features from each scale using the thresholds shown in table 1. The curve after noise reduction and smoothing is shown in fig. 6;
table 1: noise reduction threshold
6) The derivative is obtained for each point of the smooth curve,
in the formula: f (t) is the smoothed head characteristic curve;
determining a frontal sinus region by calculating the position of a point d at which the first derivative of the smoothed head characteristic curve is zero;
7) increasing 50 pixels up, down, front and back according to the d-point coordinate, and cutting the original image to obtain a subgraph only containing a frontal sinus region, as shown in fig. 7;
8) to enhance the frontal sinus region in the subgraph, wiener filtering is performed on the image, as shown in fig. 8;
9) binarizing the subgraph as shown in FIG. 9;
10) the method of zeroing the connected region connected to the edge eliminates the edge effect, and the result is shown in fig. 10;
11) calculating the mass center coordinate of the frontal sinus region, wherein the formula is as follows;
in the formula: (x)c,yc) The abscissa and ordinate of the centroid;
Iijthe gray value of the corresponding point is obtained;
FIG. 11 is a display of the centroid on two images to be registered;
12) and registering the images through the difference of the coordinates of the mass centers.
In the above embodiment, histogram normalization is performed on the image to be registered, a gradient algorithm is further applied to extract the head contour, DB5 wavelet is applied to perform frequency domain analysis on the extracted head characteristic curve, and noise reduction is performed by setting a correlation threshold for different scale frequencies, so as to obtain a smooth curve. And finding out the corresponding coordinates of the frontal sinus region by applying a method of deriving the curve, and cutting the coordinates. And calculating the centroid of the cut subgraph, and further realizing registration of the two images through the centroid coordinate.
The invention relates to a method and a device for registering head multi-mode CT images and radiotherapy equipment, which use an image recognition technology to use the biological characteristics of the human skull as the basis of image registration and use the position information of characteristic points to correct the spatial positions of a reference image and an image to be registered, thereby effectively reducing the positioning error, registering the images at the pixel level and leading the tumor part of a patient to be treated more accurately.
It should be understood that the various techniques described herein may be implemented in connection with hardware or software or, alternatively, with a combination of both. Thus, the methods and apparatus of the present invention, or certain aspects or portions thereof, may take the form of program code (i.e., instructions) embodied in tangible media, such as floppy diskettes, CD-ROMs, hard drives, or any other machine-readable storage medium, wherein, when the program is loaded into and executed by a machine, such as a computer, the machine becomes an apparatus for practicing the invention.
The above embodiments are merely illustrative of the technical concept and features of the present invention, and the purpose thereof is to enable those skilled in the art to understand the content of the present invention and implement the present invention, and not to limit the scope of the present invention, and all equivalent changes or modifications made according to the spirit of the present invention should be covered in the scope of the present invention.
Claims (9)
1. The head multi-modality CT image registration method is characterized by specifically comprising the following steps:
s1: acquiring a reference image of the head and an image to be registered of the head, and preprocessing the two acquired images before feature extraction;
s2: respectively extracting facial edges of the two preprocessed images to obtain corresponding head contour curves;
s3: carrying out quantitative analysis on the two head contour curves respectively to obtain corresponding head characteristic curves;
s4: carrying out noise reduction smoothing treatment on the two head characteristic curves;
s5: deriving each point of the two smooth head characteristic curves, wherein the point with the first derivative of zero is the frontal part of the head surface, and taking the point as the center, increasing the pixels with specified quantity respectively from top to bottom, left to right, and extracting frontal sinus regions of the two images;
s6: and calculating the mass center coordinates of the frontal sinus region of the reference image and the image to be registered, and registering the two images by taking the mass center coordinates of the frontal sinus region of the two images as the basis.
2. The head multi-modality CT image registration method of claim 1, wherein the preprocessing in S1 includes: one or more of cropping, size normalization, histogram normalization.
3. The head multi-modality CT image registration method of claim 1, wherein the establishment of the head characteristic curve in S3 specifically includes the following:
and establishing a coordinate system, wherein the vertex of the head in the head contour curve is used as an origin, the longitudinal distance of each point on the head contour curve is used as the numerical value of the x axis or the y axis, and the transverse distance of each point on the contour curve is used as the numerical value of the y axis or the x axis.
4. The head multi-modality CT image registration method of claim 1, wherein the denoising smoothing process in S4 specifically comprises the following steps:
and performing wavelet decomposition on the extracted head characteristic curve, analyzing the characteristic region frequency characteristics of the head characteristic curve, and filtering curve noise in a frequency domain to ensure that the curve is smooth.
5. Head multi-modality CT image registration apparatus, characterized by comprising:
the image acquisition module is used for acquiring a reference image of the head and an image to be registered of the head, and preprocessing the two acquired images before feature extraction;
the contour curve extraction module is used for respectively extracting the facial edges of the two preprocessed images to obtain corresponding head contour curves;
the characteristic curve acquisition module is used for respectively carrying out quantitative analysis on the two head contour curves to obtain corresponding head characteristic curves;
the noise reduction module is used for carrying out noise reduction smoothing treatment on the two head characteristic curves;
the frontal sinus region extraction module is used for deriving each point of the two smooth head characteristic curves, the point with the first derivative being zero is the frontal part of the head surface, the point is taken as the center, the pixels with the specified number are respectively added up, down, left and right, and the frontal sinus regions of the two images are extracted;
and the registration module is used for calculating the mass center coordinates of the frontal sinus region of the reference image and the image to be registered respectively, and registering the two images by taking the mass center coordinates of the frontal sinus region of the two images as the basis.
6. The head multi-modality CT image registration apparatus as claimed in claim 5, wherein the image acquisition module is capable of performing one or more pre-processing operations of cropping, size normalization, histogram normalization on the acquired reference image and the image to be registered.
7. The head multi-modality CT image registration apparatus of claim 5, wherein the feature curve acquisition module establishes the head feature curve by:
and establishing a coordinate system, wherein the vertex of the head in the head contour curve is used as an origin, the longitudinal distance of each point on the head contour curve is used as the numerical value of the x axis or the y axis, and the transverse distance of each point on the contour curve is used as the numerical value of the y axis or the x axis.
8. The head multi-modality CT image registration apparatus of claim 5, wherein the noise reduction module performs noise reduction smoothing by:
and performing wavelet decomposition on the extracted head characteristic curve, analyzing the characteristic region frequency characteristics of the head characteristic curve, and filtering curve noise in a frequency domain to ensure that the curve is smooth.
9. Radiotherapy apparatus characterized in that registration is performed using a head multi-modality CT image registration method according to any one of claims 1-4, or comprising: the head multi-modality CT image registration apparatus of any of claims 5-8.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210242421.XA CN114612533A (en) | 2022-03-11 | 2022-03-11 | Head multi-mode CT image registration method and device and radiotherapy equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210242421.XA CN114612533A (en) | 2022-03-11 | 2022-03-11 | Head multi-mode CT image registration method and device and radiotherapy equipment |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114612533A true CN114612533A (en) | 2022-06-10 |
Family
ID=81863345
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210242421.XA Pending CN114612533A (en) | 2022-03-11 | 2022-03-11 | Head multi-mode CT image registration method and device and radiotherapy equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114612533A (en) |
-
2022
- 2022-03-11 CN CN202210242421.XA patent/CN114612533A/en active Pending
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP7030050B2 (en) | Pseudo-CT generation from MR data using tissue parameter estimation | |
JP6567179B2 (en) | Pseudo CT generation from MR data using feature regression model | |
CN107545584B (en) | Method, device and system for positioning region of interest in medical image | |
US9466093B2 (en) | Anterior commissure and posterior commissure segmentation system and method | |
CN113506294B (en) | Medical image evaluation method, system, computer equipment and storage medium | |
CN110097557B (en) | Medical image automatic segmentation method and system based on 3D-UNet | |
CN112509022A (en) | Non-calibration object registration method for preoperative three-dimensional image and intraoperative perspective image | |
EP3785222B1 (en) | Systems and methods for image processing | |
Shi et al. | Many is better than one: an integration of multiple simple strategies for accurate lung segmentation in CT images | |
EP1956553B1 (en) | Visual enhancement of interval changes using a temporal subtraction technique. | |
CN106780649A (en) | The artifact minimizing technology and device of image | |
CN116993628B (en) | CT image enhancement system for tumor radio frequency ablation guidance | |
Hacihaliloglu et al. | Statistical shape model to 3D ultrasound registration for spine interventions using enhanced local phase features | |
CN116630762A (en) | Multi-mode medical image fusion method based on deep learning | |
CN116309647B (en) | Method for constructing craniocerebral lesion image segmentation model, image segmentation method and device | |
EP1956552B1 (en) | Visual enhancement of interval changes using a temporal subtraction technique | |
CN105184741A (en) | Three-dimensional CBCT (cone-beam computed tomography) image denoising method on the basis of improved nonlocal means | |
CN110652660B (en) | Patient positioning detection system | |
CN116883372A (en) | Method and system for adaptively identifying tumor based on blood vessel region image | |
CN109816665B (en) | Rapid segmentation method and device for optical coherence tomography image | |
CN114612533A (en) | Head multi-mode CT image registration method and device and radiotherapy equipment | |
CN111477304A (en) | Tumor irradiation imaging combination method for fusing PET (positron emission tomography) image and MRI (magnetic resonance imaging) image | |
CN113077433B (en) | Deep learning-based tumor target area cloud detection device, system, method and medium | |
CN116168097A (en) | Method, device, equipment and medium for constructing CBCT sketching model and sketching CBCT image | |
Al Abboodi et al. | Supervised Transfer Learning for Multi Organs 3D Segmentation With Registration Tools for Metal Artifact Reduction in CT Images |
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 |