CN112641458A - Medical image processing method, apparatus, image processing device and medium - Google Patents
Medical image processing method, apparatus, image processing device and medium Download PDFInfo
- Publication number
- CN112641458A CN112641458A CN202011511183.5A CN202011511183A CN112641458A CN 112641458 A CN112641458 A CN 112641458A CN 202011511183 A CN202011511183 A CN 202011511183A CN 112641458 A CN112641458 A CN 112641458A
- Authority
- CN
- China
- Prior art keywords
- blood flow
- preset
- artery
- blood
- image processing
- 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
- 238000012545 processing Methods 0.000 title claims abstract description 29
- 238000003672 processing method Methods 0.000 title claims abstract description 26
- 230000017531 blood circulation Effects 0.000 claims abstract description 123
- 210000001367 artery Anatomy 0.000 claims abstract description 60
- 238000012905 input function Methods 0.000 claims abstract description 54
- 238000000034 method Methods 0.000 claims abstract description 25
- 239000008280 blood Substances 0.000 claims abstract description 24
- 210000004369 blood Anatomy 0.000 claims abstract description 24
- 230000006870 function Effects 0.000 claims description 54
- 238000009826 distribution Methods 0.000 claims description 41
- 210000004556 brain Anatomy 0.000 claims description 17
- 238000012937 correction Methods 0.000 claims description 12
- 210000001715 carotid artery Anatomy 0.000 claims description 9
- 210000004731 jugular vein Anatomy 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 5
- 238000010801 machine learning Methods 0.000 claims description 4
- 230000003727 cerebral blood flow Effects 0.000 claims description 2
- 238000002591 computed tomography Methods 0.000 description 43
- 210000005013 brain tissue Anatomy 0.000 description 10
- 239000002872 contrast media Substances 0.000 description 8
- 210000004204 blood vessel Anatomy 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 239000000243 solution Substances 0.000 description 5
- 238000000354 decomposition reaction Methods 0.000 description 4
- 238000009795 derivation Methods 0.000 description 4
- 208000006011 Stroke Diseases 0.000 description 3
- 210000003462 vein Anatomy 0.000 description 3
- ZCYVEMRRCGMTRW-UHFFFAOYSA-N 7553-56-2 Chemical compound [I] ZCYVEMRRCGMTRW-UHFFFAOYSA-N 0.000 description 2
- 208000016988 Hemorrhagic Stroke Diseases 0.000 description 2
- 208000032382 Ischaemic stroke Diseases 0.000 description 2
- 238000007476 Maximum Likelihood Methods 0.000 description 2
- 230000036770 blood supply Effects 0.000 description 2
- 238000010790 dilution Methods 0.000 description 2
- 239000012895 dilution Substances 0.000 description 2
- 201000010099 disease Diseases 0.000 description 2
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 238000005534 hematocrit Methods 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 208000020658 intracerebral hemorrhage Diseases 0.000 description 2
- 229910052740 iodine Inorganic materials 0.000 description 2
- 239000011630 iodine Substances 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 206010028980 Neoplasm Diseases 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000002490 cerebral effect Effects 0.000 description 1
- 238000003759 clinical diagnosis Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000010253 intravenous injection Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 230000036962 time dependent Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000000472 traumatic effect Effects 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
- A61B6/02—Arrangements for diagnosis sequentially in different planes; Stereoscopic radiation diagnosis
- A61B6/03—Computed tomography [CT]
- A61B6/032—Transmission computed tomography [CT]
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
- A61B5/026—Measuring blood flow
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis 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
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis 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
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Medical Informatics (AREA)
- General Health & Medical Sciences (AREA)
- Veterinary Medicine (AREA)
- Biophysics (AREA)
- Physics & Mathematics (AREA)
- Public Health (AREA)
- Animal Behavior & Ethology (AREA)
- Surgery (AREA)
- Pathology (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Radiology & Medical Imaging (AREA)
- Optics & Photonics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- High Energy & Nuclear Physics (AREA)
- Physiology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Hematology (AREA)
- Cardiology (AREA)
- Pulmonology (AREA)
- Theoretical Computer Science (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
The embodiment of the invention discloses a medical image processing method, a device, image processing equipment and a medium, wherein the method comprises the following steps: determining an artery input function of an artery blood vessel for supplying blood to a target position and attenuation coefficient enhancement information of a single voxel corresponding to the artery blood vessel according to a plurality of groups of dynamic enhanced CT images acquired at different time points; and inputting the artery input function and the attenuation coefficient enhancement information into a preset estimation model to obtain preset blood flow parameters of the target part, wherein the preset blood flow parameters comprise blood flow speed and average flow time, and the preset estimation model is constructed based on a box-type attenuation function. The technical problem that the blood flow parameters of dynamic enhanced CT cannot be accurately determined in the prior art is solved.
Description
Technical Field
The embodiment of the invention relates to the field of image processing, in particular to a medical image processing method, a medical image processing device, image processing equipment and a medium.
Background
Global disease burden studies have shown that the lifetime risk of stroke in the population of china is 39.9%, the first cause of life loss due to disease, and stroke is divided into two categories, ischemic stroke and hemorrhagic stroke, wherein ischemic stroke is caused by complete or partial blockage of blood supply, and hemorrhagic stroke is caused by non-traumatic rupture of cerebral blood vessels leading to blood accumulation in the brain. Clinically, information for diagnosing stroke is usually obtained from cerebral blood flow parameters measured by dynamic enhanced CT (Computed tomography) imaging, such as blood volume (CBV), blood flow rate (CBF), mean time to flow (MTT), and Time To Peak (TTP).
Dynamic enhanced CT is an imaging method that collects and detects contrast agent circulating with blood flow in the brain after intravenous injection of iodine contrast agent. At acquisition, the not yet enhanced head is first scanned and the reconstructed image is defined as a mode image. The brain projection data acquired at a plurality of successive time points are then reconstructed into a three-dimensional brain dynamic enhanced CT image corresponding to each time point. The dynamically enhanced series of images may be obtained by subtracting a model image from later acquired images at each time point, which contains blood flow parameter information. As known from index dilution theory, the attenuation coefficient enhancement information of iodine contrast agent for each voxel in brain tissue can be expressed as a convolution of the residual function of blood flow extension and the arterial input function. If both the enhancement information and the arterial input function can be measured by dynamically enhancing the CT image, the blood flow extension residual function can be derived according to the index dilution theory, and the blood flow extension residual function can be decomposed into blood flow parameters of individual voxels. From the viewpoint of mathematical model, this is a process of solving an inverse problem, and the conventional method for solving this inverse problem is based on Singular Value Decomposition (SVD) and a series of methods for improving it. However, since dynamic enhanced CT is limited by radiation dose limits or actual scanning conditions, often with lower image quality than normal CT, the inverse problem to be solved is often an ill-posed problem, i.e. the derived blood flow parameters may be unstable or inaccurate.
In conclusion, the prior art has the technical problem that the blood flow parameters of the dynamic enhanced CT cannot be accurately determined.
Disclosure of Invention
The embodiment of the invention provides a medical image processing method, a medical image processing device, image processing equipment and a medical image processing medium, and solves the technical problem that the blood flow parameters of dynamic enhanced CT cannot be accurately determined in the prior art.
In a first aspect, an embodiment of the present invention provides a medical image processing method, where the method includes:
determining an artery input function of an artery blood vessel for supplying blood to a target position and attenuation coefficient enhancement information of a single voxel corresponding to the artery blood vessel according to a plurality of groups of dynamic enhanced CT images acquired at different time points;
and inputting the artery input function and the attenuation coefficient enhancement information into a preset estimation model to obtain preset blood flow parameters of the target part, wherein the preset blood flow parameters comprise blood flow speed and average flow time, and the preset estimation model is constructed based on a box-type attenuation function.
In a second aspect, an embodiment of the present invention further provides a medical image processing apparatus, including:
the acquisition module is used for determining an artery input function of an artery blood vessel for supplying blood to a target part and attenuation coefficient enhancement information of a single voxel corresponding to the artery blood vessel according to a plurality of groups of dynamic enhanced CT images acquired at different time points;
and the blood flow parameter determination module is used for inputting the artery input function and the attenuation coefficient enhancement information into a preset estimation model to obtain preset blood flow parameters of the target part, wherein the preset blood flow parameters comprise blood flow speed and average flow time, and the preset estimation model is constructed based on a box-type attenuation function.
In a third aspect, an embodiment of the present invention further provides an image processing apparatus, including:
one or more processors;
storage means for storing one or more programs;
when executed by the one or more processors, cause the one or more processors to implement a medical image processing method as in any embodiment.
In a fourth aspect, embodiments of the present invention further provide a storage medium containing computer-executable instructions, which when executed by a computer processor, are configured to perform the medical image processing method according to any of the embodiments.
The technical scheme of the medical image processing method provided by the embodiment of the invention comprises the following steps: determining an artery input function of an artery blood vessel for supplying blood to a target part and attenuation coefficient enhancement information of a single voxel corresponding to the artery blood vessel according to a plurality of groups of dynamic enhanced CT images acquired at different time points; and inputting the artery input function and the attenuation coefficient enhancement information into a preset estimation model to obtain preset blood flow parameters of the target part, wherein the preset blood flow parameters comprise blood flow speed and average flow time, and the preset estimation model is constructed based on the box-type attenuation function. Compared with an ideal physiological model required in singular value decomposition, the preset estimation model constructed based on the box-type attenuation function is more in line with the actual situation, so that the accuracy of the preset blood flow parameters determined based on the preset estimation model is higher.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a flowchart of a medical image processing method according to an embodiment of the present invention;
FIG. 2 is a flowchart of a medical image processing method according to a second embodiment of the present invention;
fig. 3 is a block diagram of a medical image processing apparatus according to a third embodiment of the present invention;
fig. 4 is a block diagram of a medical image processing apparatus according to a third embodiment of the present invention;
fig. 5 is a block diagram of an image processing apparatus according to a fourth embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described through embodiments with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. 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.
Example one
Fig. 1 is a flowchart of a medical image processing method according to an embodiment of the present invention. The technical scheme of the embodiment is suitable for the condition that the preset blood flow parameters corresponding to the dynamic enhanced CT image are determined by the preset estimation model constructed based on the box-type attenuation function. The method can be executed by a medical image processing device provided by the embodiment of the invention, and the device can be realized in a software and/or hardware manner and is configured to be applied in a processor of an image processing device. The method specifically comprises the following steps:
s101, determining an artery input function of an artery blood vessel for supplying blood to a target part and attenuation coefficient enhancement information of a single voxel corresponding to the artery blood vessel according to a plurality of groups of dynamic enhanced CT images acquired at different time points.
The target site is a site to be analyzed for a blood flow rate parameter, preferably but not limited to a brain, a target tumor region, and the like, and the technical scheme is explained by taking the brain as an example in this embodiment.
It will be appreciated that if the target site is the brain, then the dynamic enhanced CT image is a dynamic enhanced CT image of the brain, and the arterial blood vessels supplying the brain tissue are the carotid arteries.
Wherein the different time points refer to the image acquisition time points after the contrast agent enters the brain tissue. The timing and number of image acquisition time points can be set according to the needs of the clinician. The present embodiment uses the reconstructed image corresponding to the scan data acquired at each time point as a set of dynamic enhanced CT images.
The format of the dynamic enhanced CT image is DICOM format.
The Arterial Input Function (AIF) describes, among other things, the temporal behavior of contrast agents through the imaged voxels. It will be appreciated that in practice the input concentration-time curves for each voxel are not exactly the same, and therefore the intensity variation information of the contrast agent for each voxel is different, and therefore only the attenuation coefficient enhancement information of the contrast agent for a single voxel of the blood supply vessel can be estimated. The present embodiment does not limit the estimation method of the attenuation coefficient enhancement information of a single voxel, and the attenuation coefficient enhancement information of the single voxel may be determined by using the prior art.
The method for determining the artery input function comprises the following steps: determining a pixel mean value in a carotid artery region of a plurality of groups of brain dynamic enhanced CT images acquired at different time points, taking a relative enhancement degree corresponding to the pixel mean value as a first relative enhancement degree, determining a pixel mean value in a jugular vein region of the plurality of groups of brain dynamic enhanced CT images, and taking a relative enhancement degree corresponding to the pixel mean value as a second relative enhancement degree; constructing an arterial input function curve corresponding to the first relative enhancement degree and a venous output function curve corresponding to the second relative enhancement degree by taking the acquisition time as a horizontal axis and taking the relative enhancement degree as a vertical axis; completing the correction of the arterial input function curve according to the curve area corresponding to the venous output function curve so as to update the arterial input function curve; and determining the artery input function according to the updated artery input function curve. Because the blood outflow volume of the vein blood vessel is the same as the blood inflow volume of the artery blood vessel, the curve area corresponding to the artery input function is the same as the curve area corresponding to the vein output function, and therefore the artery input function curve can be corrected according to the curve area corresponding to the vein output function curve, the problem that the artery input function is inaccurate due to low pixel value of the artery region on the dynamic enhanced CT image is solved, and the accuracy of the artery input function is improved.
In some embodiments, the method of determining a carotid artery region comprises: adding corresponding frames of a plurality of groups of dynamic enhanced CT images acquired at different time points to obtain a first image, then segmenting the first image by using a first preset threshold value to obtain a carotid artery template, and determining a carotid artery region of the dynamic enhanced CT image acquired at each time point by using the carotid artery template; correspondingly, the first image is segmented by using a second preset threshold value to obtain a jugular vein template, and the jugular vein template is used for determining a jugular vein area of the dynamic enhanced CT image acquired at each time point.
It is understood that the dynamically enhanced CT image is a pre-processed CT image. The preprocessing methods include, but are not limited to, filtering denoising processing and motion correction processing. The motion correction process includes intra-frame motion correction and inter-frame motion correction. In this embodiment, preferably, the intra-frame motion correction processing is performed on the plurality of groups of dynamic enhanced CT images after the filtering and denoising processing, so as to update the dynamic enhanced CT images; and selecting a reference frame image from the updated multiple groups of dynamic enhanced CT images, and then carrying out rigid registration on other frame images according to the reference frame image so as to update the multiple groups of dynamic enhanced CT images again. It can be understood that the dynamically enhanced CT image after the intra-frame motion correction and the inter-frame motion correction has no motion artifact or only less motion artifact, and can avoid the situation that the accuracy of the blood input function and the attenuation coefficient enhancement information of the single voxel is low due to the motion artifact.
S102, inputting the artery input function and the attenuation coefficient enhancement information into a preset estimation model to obtain preset blood flow parameters of the target part, wherein the preset blood flow parameters comprise blood flow speed and average flow time, and the preset estimation model is constructed based on a box-type attenuation function.
After the artery input function and the attenuation coefficient enhancement information of a single voxel are obtained, the artery input function and the attenuation coefficient enhancement information are input into a preset estimation model constructed on the basis of a box-type attenuation function to obtain preset blood flow parameters. The preset blood flow parameters include blood flow rate and mean flow time. The blood flow velocity is the linear velocity of brain blood flowing in a blood vessel, namely the distance of a particle advancing in the blood vessel in unit time; by average flow-through time is meant the average time the contrast agent stays in the tissue.
The method for constructing the preset estimation model comprises the following steps:
the time-dependent variation curve of the attenuation coefficient enhancement information of a single voxel can be expressed as:
where μ is the blood flow parameter to be estimated, h (μ, t) is the scaled residual function of the blood flow, Ca(t) is the Arterial Input Function (AIF) and ε (t) is noise. The scaled residual function of blood flow can be further written as a function of blood flow rate (CBF), tissue density ρ, and residual function r (t), as follows:
h(μ,t)＝CBF·ρ·r(t) (2)
where the set μ includes the blood flow rate (CBF) and the mean flow time (MTT). Constructing r (t) as a function based on a box-type decay function, then one can obtain from the above equation:
wherein, T00.632 × MTT. It will be appreciated that the convolution in equation (1) can be expanded to sample the integral of the arterial input function as follows:
it is understood that substituting equation (3) into equation (4) results in h (μ, t) by deconvolution, thus resulting in CBF and MTT. After CBF and MTT are obtained, the product of the two can be calculated to obtain the blood volume (CBV). In addition, the abscissa corresponding to the maximum value of y (t) in the formula (4) may be determined, and the time corresponding to the abscissa is the time To Top (TTP). All the blood flow parameters commonly used in clinic can be obtained.
The technical scheme of the medical image processing method provided by the embodiment of the invention comprises the following steps: determining an artery input function of an artery blood vessel for supplying blood to a target part and attenuation coefficient enhancement information of a single voxel corresponding to the artery blood vessel according to a plurality of groups of dynamic enhanced CT images acquired at different time points; and inputting the artery input function and the attenuation coefficient enhancement information into a preset estimation model to obtain preset blood flow parameters of the target part, wherein the preset blood flow parameters comprise blood flow speed and average flow time, and the preset estimation model is constructed based on the box-type attenuation function. Compared with an ideal physiological model required in singular value decomposition, the preset estimation model constructed based on the box-type attenuation function is more in line with the actual situation, so that the accuracy of the preset blood flow parameters determined based on the preset estimation model is higher.
Example two
Fig. 2 is a flowchart of a medical image processing method according to a second embodiment of the present invention. On the basis of the embodiment, the embodiment of the invention adds a step of estimating the preset blood flow parameter distribution.
Correspondingly, the method of the embodiment comprises the following steps:
s201, according to a plurality of groups of dynamic enhanced CT images acquired at different time points, determining an artery input function of an artery blood vessel for supplying blood to a target position and attenuation coefficient enhancement information of a single voxel corresponding to the artery blood vessel.
S202, inputting the artery input function and the attenuation coefficient enhancement information into a preset estimation model to obtain preset blood flow parameters of the target part, wherein the preset blood flow parameters comprise blood flow speed and average flow time, and the preset estimation model is constructed based on a box-type attenuation function.
S203, determining an updating function of the mean value and the variance of the preset blood flow parameters determined by the preset estimation model by a Bayesian machine learning method.
On the basis of the foregoing embodiments, the present embodiment further performs derivation estimation on the preset blood flow parameters by a bayesian machine learning method to determine the distribution of the preset blood flow parameters.
Since y (t) corresponds to N discrete sampling time points in the scan, the noise epsilon (t) can be modeled as a zero-mean gaussian distribution at each acquisition time point, and each image frame is independent in time, equation (4) can be expressed by the following discrete expression:
where i is the identification of the image frame, ωiProportional to the weight of the detection event (i.e., the acquisition duration of the three-dimensional enhanced CT image of the brain identified as i). Constructing the data mismatch terms according to equation (4), then the derivation problem becomes one of finding the parameter μ that maximizes the log-likelihood function, as follows:
because omega is in the condition that the acquisition time of each three-dimensional dynamic enhanced CT image is the samei1, hence the above optimization problem
Equivalent to a non-linear two-times fitting problem.
Equation (6) can be extended to a complete bayesian derivation problem that provides an overall probability distribution of the preset blood flow parameters instead of just the maximum likelihood estimates, directly reflecting uncertainty information for each parameter measurement. According to Bayesian theorem, the posterior distribution of the estimated parameter mu relative to the attenuation coefficient enhancement information, namely Y is as follows:
where P (Y | μ) is a likelihood function, P (μ) is a prior probability of the estimation parameter μ, and P (Y) is a prior probability of the attenuation coefficient enhancement information Y. It will be appreciated that the fundamental difference between this equation and equation (6) is that μ no longer represents the maximum likelihood estimate but the overall probability distribution. Estimating μ in a bayesian framework becomes a derivation problem, essentially maximizing P (μ | Y) based on given attenuation coefficient enhancement information (Y). Since it is difficult to directly calculate the posterior distribution, this embodiment uses the method of variational bayes for approximation. The basic idea of this variational bayes method is to find an analytic distribution q (μ | Y) that is close to the posterior distribution P (μ | Y) so that the relative entropy of the two is minimized. The logarithmic evidence of the distribution model of equation (7) can be written as:
wherein P (Y, μ) is a combination distribution, E*Is expected for q (. mu. | Y). ELBO is the lower limit of evidence and KL is the divergence of the approximate distributions q (μ | Y) and P (Y, μ). Since the relative entropy is always positive, ELBO provides the lower bound of the log-likelihood function. We can then find μ that brings the analytical distribution closer to the true posterior distribution by maximizing ELBO as follows:
the relative entropy in parentheses of formula (9) is always positive as well, and the optimal solution for μ can be obtained by the equivalent numerator and denominator.
For the molecule, it is ensured that it is easy to handle. Q (μ | Y) is decomposed using variational bayes, assuming that the true posterior distribution of each parameter is a multi-parameter gaussian. Such an approximate posterior distribution can be expressed as:
where k is 1, mu1Corresponding to the blood flow velocity, σ1Is the variance of the blood flow velocity distribution, m1Mean of blood flow velocity distribution; when k is 2, mu2Corresponding to the blood flow velocity, σ2Is the variance of the blood flow velocity distribution, m2Mean of blood flow velocity distribution; mu is a matrix containing the blood flow rate to be estimated and the mean flow-through time, and sigma is a matrix containing the variance sigma of the blood flow rate1And variance σ of mean flow time2M is a matrix containing the mean value m of the blood flow velocity1And average flow time m2Is measured.
For the denominator, the likelihood function of equation (6) is inserted, while the conjugate distribution, i.e. the multivariate gaussian distribution, is selected as the prior distribution, the mean of which is m0Variance is σ0. The denominator can thus be expressed as:
to ensure tractability of the denominator and to be applicable to equation (4), g (μ) can be expressed by a first order Taylor expansion of the posterior distribution as follows:
wherein J is a Jacobian matrix. After applying this linear transformation, equation (11) becomes:
the update function of the mean and variance of μ is obtained by equating equation (10) and equation (13) as follows:
as can be appreciated, σ0And m0Respectively, to the initial values of σ and m.
S204, determining the distribution of the preset blood flow parameters according to the updating function of the variance and the mean of the preset blood flow parameters.
Since μ includes the blood flow rate and the mean flow time, after the update function of the mean and the variance of μ is determined, the update function of the mean and the variance of the blood flow rate (CBF) and the update function of the mean and the variance of the mean flow time (MTT) can be determined, and the distribution of the blood flow rate is determined from the update function of the variance and the mean of the blood flow rate and the distribution of the mean flow time is determined from the update function of the variance and the mean of the mean flow time. The variance of the blood flow rate includes uncertainty information of the blood flow rate, and the variance of the average flow time includes uncertainty information of the average flow time.
After determining the distribution of the blood flow rate and the distribution of the mean flow time, the present embodiment preferably further calculates the ratio of the variance of the blood flow rate to the mean to obtain a first coefficient of variation representing the uncertainty of the blood flow rate, and calculates the ratio of the variance of the mean flow time to the mean to obtain a second coefficient of variation representing the uncertainty of the mean flow time. It will be appreciated that the greater the coefficient of variation, the greater the uncertainty in the corresponding parameter, and correspondingly, the greater the uncertainty in the dynamically enhanced CT image scan.
It will be appreciated that the blood flow rate distribution estimated by the present embodiment is only representative of the blood flow rate distribution of the capillaries within the brain tissue. In clinical diagnosis, the blood flow velocity distribution outside the brain tissue needs to be referred to, and therefore, after the blood flow velocity distribution inside the brain tissue is obtained, the blood flow velocity distribution outside the brain tissue is determined through hematocrit correction, which specifically includes: the mean value of the blood flow velocity is multiplied by a preset coefficient, and the variance of the blood flow velocity is multiplied by the square of the preset coefficient to obtain the blood flow velocity distribution outside the brain tissue. Wherein the preset coefficient is optionally 0.733.
In one embodiment, after obtaining the blood volume, the blood volume outside the brain tissue is determined by hematocrit correction, specifically: the blood volume is multiplied by a preset coefficient to obtain the blood volume outside the brain tissue. Wherein the preset coefficient is optionally 0.733.
The technical scheme of the medical image processing method provided by the embodiment of the invention can determine the preset blood flow parameters through the preset estimation model, can estimate the preset estimation model according to the variational Bayes to determine the distribution of the preset blood flow parameters, and can determine the reliability of the preset blood flow parameters determined by the dynamic enhanced CT and the accuracy of the scanning protocol setting or compare the advantages and disadvantages of different image processing methods through the distribution of the preset blood flow parameters.
EXAMPLE III
Fig. 3 is a block diagram of a medical image processing apparatus according to an embodiment of the present invention. The apparatus is used for executing the medical image processing method provided by any of the above embodiments, and the apparatus can be implemented by software or hardware. The device includes:
an obtaining module 11, configured to determine, according to multiple sets of dynamically enhanced CT images acquired at different time points, an arterial input function of an arterial blood vessel for supplying blood to a target region and attenuation coefficient enhancement information of a single voxel corresponding to the arterial blood vessel;
the blood flow parameter determining module 12 is configured to input the artery input function and the attenuation coefficient enhancement information into a preset estimation model to obtain preset blood flow parameters of the target portion, where the preset blood flow parameters include blood flow rate and average flow time, and the preset estimation model is constructed based on a box-type attenuation function.
As shown in fig. 4, the apparatus further comprises an input module 10, wherein the input module 10 is configured to: determining a pixel mean value in a carotid artery region of a plurality of groups of brain dynamic enhanced CT images acquired at different time points, taking a relative enhancement degree corresponding to the pixel mean value as a first relative enhancement degree, determining a pixel mean value in a jugular vein region of the plurality of groups of brain dynamic enhanced CT images, and taking a relative enhancement degree corresponding to the pixel mean value as a second relative enhancement degree; constructing an arterial input function curve corresponding to the first relative enhancement degree and a venous output function curve corresponding to the second relative enhancement degree by taking the acquisition time as a horizontal axis and taking the relative enhancement degree as a vertical axis; completing the correction of the arterial input function curve according to the curve area corresponding to the venous output function curve so as to update the arterial input function curve; and determining the artery input function according to the updated artery input function curve.
As shown in fig. 4, the apparatus further includes an estimation module 13, where the estimation module 13 is configured to determine an update function of the mean and the variance of the preset blood flow parameter determined by the preset estimation model by a bayesian machine learning method; and determining the distribution of the preset blood flow parameters according to the updating function of the variance and the mean of the preset blood flow parameters. The reliability of the preset blood flow parameters determined by the dynamic enhanced CT and the accuracy of the scanning protocol setting can be determined through the distribution of the preset blood flow parameters, or the advantages and disadvantages of different image processing methods can be compared.
Optionally, the estimation module 13 is further configured to determine a first coefficient of variation representing uncertainty of the blood flow rate according to the variance and the mean of the blood flow rate; a second coefficient of variation representing uncertainty of the mean flow-through time is determined from the variance and the mean of the mean flow-through time.
According to the technical scheme of the medical image processing device provided by the embodiment of the invention, the artery input function of the artery blood vessel used for supplying blood to the target part and the attenuation coefficient enhancement information of a single voxel corresponding to the artery blood vessel are determined by the acquisition module according to a plurality of groups of dynamic enhanced CT images acquired at different time points; the artery input function and the attenuation coefficient enhancement information are input into a preset estimation model through a blood flow parameter determination module to obtain preset blood flow parameters of the target part, the preset blood flow parameters comprise blood flow speed and average flow time, and the preset estimation model is constructed based on a box-type attenuation function. Compared with an ideal physiological model required in singular value decomposition, the preset estimation model constructed based on the box-type attenuation function is more in line with the actual situation, so that the accuracy of the preset blood flow parameters determined based on the preset estimation model is higher.
The medical image processing device provided by the embodiment of the invention can execute the medical image processing method provided by any embodiment of the invention, and has corresponding functional modules and beneficial effects of the execution method.
Example four
Fig. 5 is a schematic structural diagram of an image processing apparatus according to a fourth embodiment of the present invention, as shown in fig. 5, the apparatus includes a processor 201, a memory 202, an input device 203, and an output device 204; the number of the processors 201 in the device may be one or more, and one processor 201 is taken as an example in fig. 5; the processor 201, the memory 202, the input device 203 and the output device 204 in the apparatus may be connected by a bus or other means, and fig. 5 illustrates the connection by a bus as an example.
The memory 202, as a computer-readable storage medium, may be used for storing software programs, computer-executable programs, and modules, such as program instructions/modules (e.g., the acquisition module 11 and the blood flow parameter determination module 12) corresponding to the medical image processing method in the embodiment of the present invention. The processor 201 executes various functional applications of the device and data processing by executing software programs, instructions and modules stored in the memory 202, that is, implements the medical image processing method described above.
The memory 202 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created according to the use of the terminal, and the like. Further, the memory 202 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some examples, the memory 202 may further include memory located remotely from the processor 201, which may be connected to the device over a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The input device 203 may be used to receive input numeric or character information and generate key signal inputs related to user settings and function controls of the apparatus.
The output means 204 may comprise a display device such as a display screen, for example, a display screen of a user terminal, at least for outputting the preset blood flow parameter and/or the distribution of the preset blood flow parameter.
EXAMPLE five
Embodiments of the present invention also provide a storage medium containing computer-executable instructions, which when executed by a computer processor, are configured to perform a method of medical image processing, the method comprising:
determining an artery input function of an artery blood vessel for supplying blood to a target position and attenuation coefficient enhancement information of a single voxel corresponding to the artery blood vessel according to a plurality of groups of dynamic enhanced CT images acquired at different time points;
and inputting the artery input function and the attenuation coefficient enhancement information into a preset estimation model to obtain preset blood flow parameters of the target part, wherein the preset blood flow parameters comprise blood flow speed and average flow time, and the preset estimation model is constructed based on a box-type attenuation function.
Of course, the storage medium containing the computer-executable instructions provided by the embodiments of the present invention is not limited to the method operations described above, and may also perform related operations in the medical image processing method provided by any embodiment of the present invention.
From the above description of the embodiments, it is obvious for those skilled in the art that the present invention can be implemented by software and necessary general hardware, and certainly, can also be implemented by hardware, but the former is a better embodiment in many cases. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which may be stored in a computer-readable storage medium, such as a floppy disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a FLASH Memory (FLASH), a hard disk or an optical disk of a computer, and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network device) to execute the medical image processing method according to the embodiments of the present invention.
It should be noted that, in the embodiment of the medical image processing apparatus, the units and modules included in the embodiment are only divided according to the functional logic, but are not limited to the above division as long as the corresponding functions can be realized; in addition, specific names of the functional units are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present invention.
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims (10)
1. A medical image processing method, characterized by comprising:
determining an artery input function of an artery blood vessel for supplying blood to a target position and attenuation coefficient enhancement information of a single voxel corresponding to the artery blood vessel according to a plurality of groups of dynamic enhanced CT images acquired at different time points;
and inputting the artery input function and the attenuation coefficient enhancement information into a preset estimation model to obtain preset blood flow parameters of the target part, wherein the preset blood flow parameters comprise blood flow speed and average flow time, and the preset estimation model is constructed based on a box-type attenuation function.
2. The method of claim 1, wherein the target site is a brain and the arterial vessel is a carotid artery, and wherein the determining the arterial input function comprises:
determining a pixel mean value in a carotid artery region of a plurality of groups of brain dynamic enhanced CT images acquired at different time points, taking a relative enhancement degree corresponding to the pixel mean value as a first relative enhancement degree, determining a pixel mean value in a jugular vein region of the plurality of groups of brain dynamic enhanced CT images, and taking a relative enhancement degree corresponding to the pixel mean value as a second relative enhancement degree;
constructing an arterial input function curve corresponding to the first relative enhancement degree and a venous output function curve corresponding to the second relative enhancement degree by taking the acquisition time as a horizontal axis and taking the relative enhancement degree as a vertical axis;
finishing the correction of the arterial input function curve according to the curve area corresponding to the venous output function curve so as to update the arterial input function curve;
and determining the artery input function according to the updated artery input function curve.
3. The method of claim 1, wherein the mathematical expression of the predetermined estimation model comprises:
where μ is the cerebral blood flow parameter to be estimated, h (μ, T) is the scaled residual function of blood flow, CBF is the blood flow rate, MTT is the mean flow time, T0T is the acquisition time of the dynamic enhanced CT image, and is 0.632 multiplied by MTT; y (t) attenuation coefficient enhancement information for individual voxels, Ca(t) is the arterial input function.
4. The method of claim 3, further comprising:
calculating the product of the blood flow rate and the average flow-through time as the blood volume.
5. The method of claim 3, further comprising:
determining an updating function of the mean value and the variance of the preset blood flow parameters determined by the preset estimation model by a Bayesian machine learning method;
and determining the distribution of the preset blood flow parameters according to the updating function of the variance and the mean of the preset blood flow parameters.
6. The method of claim 5, wherein the update function is as follows:
wherein σnewContaining the variance of the blood flow velocity and the variance of the mean flow time, mnewComprising the mean value of the blood flow rate and the mean value of the mean flow time, gi(mu) is a discrete expression of formula (1), gi(mu) a first order Taylor expansion comprising gi(m) and the Jacobian matrix Ji，σ0And m0Initial values of σ and m, ωiIs proportional to the acquisition duration, y, of the three-dimensional dynamically enhanced CT image identified as iiThe information is enhanced for the attenuation coefficient corresponding to the image frame identified as i.
7. The method of claim 6, further comprising:
determining a first coefficient of variation representing uncertainty of the blood flow rate according to the variance and the mean of the blood flow rate;
and determining a second coefficient of variation representing uncertainty of the average flow-through time according to the variance and the mean of the average flow-through time.
8. A medical image processing apparatus, comprising:
the acquisition module is used for determining an artery input function of an artery blood vessel for supplying blood to a target part and attenuation coefficient enhancement information of a single voxel corresponding to the artery blood vessel according to a plurality of groups of dynamic enhanced CT images acquired at different time points;
and the blood flow parameter determination module is used for inputting the artery input function and the attenuation coefficient enhancement information into a preset estimation model to obtain preset blood flow parameters of the target part, wherein the preset blood flow parameters comprise blood flow speed and average flow time, and the preset estimation model is constructed based on a box-type attenuation function.
9. An image processing apparatus characterized by comprising:
one or more processors;
storage means for storing one or more programs;
when executed by the one or more processors, cause the one or more processors to implement the medical image processing method of any one of claims 1-7.
10. A storage medium containing computer-executable instructions for performing the medical image processing method of any one of claims 1-7 when executed by a computer processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011511183.5A CN112641458A (en) | 2020-12-18 | 2020-12-18 | Medical image processing method, apparatus, image processing device and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011511183.5A CN112641458A (en) | 2020-12-18 | 2020-12-18 | Medical image processing method, apparatus, image processing device and medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112641458A true CN112641458A (en) | 2021-04-13 |
Family
ID=75358399
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011511183.5A Pending CN112641458A (en) | 2020-12-18 | 2020-12-18 | Medical image processing method, apparatus, image processing device and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112641458A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113100803A (en) * | 2021-04-20 | 2021-07-13 | 西门子数字医疗科技(上海)有限公司 | Method, apparatus, computer device and medium for displaying venous thrombosis |
-
2020
- 2020-12-18 CN CN202011511183.5A patent/CN112641458A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113100803A (en) * | 2021-04-20 | 2021-07-13 | 西门子数字医疗科技(上海)有限公司 | Method, apparatus, computer device and medium for displaying venous thrombosis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10580526B2 (en) | System and method for calculating vessel flow parameters based on angiography | |
| EP3473184A1 (en) | Method and system for sensitivity analysis in modeling blood flow characteristics | |
| EP1402475B1 (en) | Analysis of successive data sets | |
| JP6987875B2 (en) | Methods and systems for modeling the human heart and atrium | |
| US9095309B2 (en) | Method and apparatus for quantifying the behavior of an administered contrast agent | |
| CN109389653B (en) | Cardiac image reconstruction method, cardiac image reconstruction device, computer equipment and readable storage medium | |
| CN109377481B (en) | Image quality evaluation method, image quality evaluation device, computer equipment and storage medium | |
| CN110136076B (en) | Medical scanning imaging method, device, storage medium and computer equipment | |
| JP2006500099A (en) | Display image data information | |
| CN111753947A (en) | Resting brain network construction method, device, equipment and computer storage medium | |
| Marin et al. | Numerical surrogates for human observers in myocardial motion evaluation from SPECT images | |
| CN112641458A (en) | Medical image processing method, apparatus, image processing device and medium | |
| CA3050591A1 (en) | Myocardial blood flow with reliability feature | |
| Fang et al. | Improving low-dose blood–brain barrier permeability quantification using sparse high-dose induced prior for patlak model | |
| CN109345606B (en) | Cardiac image reconstruction method, cardiac image reconstruction device, computer equipment and readable storage medium | |
| JP5081224B2 (en) | A method to distinguish motion parameter estimates for dynamic molecular imaging procedures | |
| Marin et al. | Deformable left‐ventricle mesh model for motion‐compensated filtering in cardiac gated SPECT | |
| Levi et al. | Comparison of automated beam hardening correction (ABHC) algorithms for myocardial perfusion imaging using computed tomography | |
| Lehnert et al. | Large-scale Bayesian spatial-temporal regression with application to cardiac MR-perfusion imaging | |
| CN116091444A (en) | Side branch evaluation method and device, storage medium and terminal | |
| CN114209344A (en) | Collateral circulation state evaluation method and device, storage medium and electronic equipment | |
| CN110929604A (en) | Screening method, device and system based on flow velocity of contrast image and storage medium | |
| GB2502817B (en) | Equalising smoothing for scan comparison to database | |
| Vashistha et al. | Non-invasive arterial input function estimation using an MRI atlas and machine learning | |
| US11704795B2 (en) | Quality-driven image processing |
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 |