CN117689582A - Correction method for motion artifact in image, electronic device and storage medium - Google Patents
Correction method for motion artifact in image, electronic device and storage medium Download PDFInfo
- Publication number
- CN117689582A CN117689582A CN202211041644.6A CN202211041644A CN117689582A CN 117689582 A CN117689582 A CN 117689582A CN 202211041644 A CN202211041644 A CN 202211041644A CN 117689582 A CN117689582 A CN 117689582A
- Authority
- CN
- China
- Prior art keywords
- image
- corrected
- patient
- artifact
- examination
- 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
- 238000012937 correction Methods 0.000 title claims abstract description 138
- 238000000034 method Methods 0.000 title claims abstract description 63
- 238000007689 inspection Methods 0.000 claims abstract description 57
- 238000012545 processing Methods 0.000 claims abstract description 32
- 230000008859 change Effects 0.000 claims description 40
- 238000012549 training Methods 0.000 claims description 28
- 238000001914 filtration Methods 0.000 claims description 27
- 238000013528 artificial neural network Methods 0.000 claims description 23
- 238000006073 displacement reaction Methods 0.000 claims description 19
- 238000004422 calculation algorithm Methods 0.000 claims description 11
- 238000004590 computer program Methods 0.000 claims description 10
- 238000013527 convolutional neural network Methods 0.000 claims description 10
- 238000012216 screening Methods 0.000 claims description 9
- 230000003287 optical effect Effects 0.000 claims description 7
- 201000010099 disease Diseases 0.000 abstract 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 abstract 1
- 230000000875 corresponding effect Effects 0.000 description 75
- 238000004364 calculation method Methods 0.000 description 15
- 230000008569 process Effects 0.000 description 14
- 238000001514 detection method Methods 0.000 description 8
- 238000002591 computed tomography Methods 0.000 description 7
- 238000010276 construction Methods 0.000 description 6
- 230000006870 function Effects 0.000 description 6
- 238000005481 NMR spectroscopy Methods 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 238000003384 imaging method Methods 0.000 description 4
- 238000012806 monitoring device Methods 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 206010011224 Cough Diseases 0.000 description 2
- 208000012661 Dyskinesia Diseases 0.000 description 2
- 208000015592 Involuntary movements Diseases 0.000 description 2
- 230000017531 blood circulation Effects 0.000 description 2
- 210000004204 blood vessel Anatomy 0.000 description 2
- 210000001175 cerebrospinal fluid Anatomy 0.000 description 2
- 230000002596 correlated effect Effects 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 239000003814 drug Substances 0.000 description 2
- 238000009499 grossing Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000017311 musculoskeletal movement, spinal reflex action Effects 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 238000001208 nuclear magnetic resonance pulse sequence Methods 0.000 description 2
- 230000000737 periodic effect Effects 0.000 description 2
- 230000010349 pulsation Effects 0.000 description 2
- 230000000241 respiratory effect Effects 0.000 description 2
- 238000000264 spin echo pulse sequence Methods 0.000 description 2
- 230000001629 suppression Effects 0.000 description 2
- 230000009747 swallowing Effects 0.000 description 2
- 238000002604 ultrasonography Methods 0.000 description 2
- 230000021542 voluntary musculoskeletal movement Effects 0.000 description 2
- 239000002699 waste material Substances 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000003702 image correction Methods 0.000 description 1
- 238000002600 positron emission tomography Methods 0.000 description 1
Landscapes
- Magnetic Resonance Imaging Apparatus (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
The invention discloses a correction method of motion artifact in an image, electronic equipment and a storage medium, wherein the correction method comprises the following steps: acquiring an image to be corrected of a current patient; acquiring offset parameters corresponding to motion artifacts generated by movement of a current patient in a target examination scene based on an image to be corrected; and according to the target inspection scene and the offset parameter, performing artifact correction processing on the image to be corrected by adopting a matched preset artifact correction network, and obtaining a target image of the current patient. According to the method, the image to be corrected, which contains the motion artifact, is arbitrarily identified, the preset artifact correction network matched with the image to be corrected is timely identified, so that the correction operation of removing the artifact of the current image to be corrected is achieved, the image containing the motion artifact is optimized, the motion artifact is eliminated, the image without the motion artifact is finally displayed, the image quality is effectively improved, and the follow-up doctor and other staff can conveniently determine the disease precision of the patient.
Description
Technical Field
The present invention relates to the field of image processing technologies, and in particular, to a method for correcting motion artifacts in an image, an electronic device, and a storage medium.
Background
In magnetic resonance examinations, the patient needs to lie still on the examination couch in order to cooperate with the completion of the whole magnetic resonance examination item. If the patient moves during the examination, motion artifacts are created on the reconstructed image; sometimes, because the examination time is long, the patient is unavoidable to generate position movement, so that motion artifact is caused; in addition, it is difficult for old people and children to remain stationary on the hospital bed for a certain period of time, so that motion artifacts are easily generated in the examination.
At present, although some modes for processing image artifacts also appear, the problems that the processing steps are complex, the processing efficiency, the processing quality and the like do not meet the actual image processing requirements exist.
Disclosure of Invention
The invention aims to overcome the defect that an image artifact removal scheme in the prior art is not accurate enough and cannot meet actual scene requirements, and provides a correction method of motion artifacts in an image, electronic equipment and a storage medium.
The invention solves the technical problems by the following technical scheme:
the invention provides a correction method of motion artifact in an image, which comprises the following steps:
acquiring an image to be corrected of a current patient;
Acquiring offset parameters corresponding to motion artifacts generated by movement of the current patient in the target examination scene based on the image to be corrected;
and according to the target examination scene and the offset parameter, performing artifact correction processing on the image to be corrected by adopting a matched preset artifact correction network, and acquiring a target image of the current patient.
Preferably, before the step of acquiring the image to be corrected of the current patient, the method further includes:
acquiring first images of different patients under different preset deflection angles under different types of examination scenes; wherein each first image is marked with an inspection scene;
acquiring a first image parameter corresponding to the first image;
calculating a first pixel offset of an image pixel corresponding to the motion artifact in the first image based on the first image parameter;
training a preset neural network based on the first pixel offset, and constructing a corresponding preset artifact correction network.
Preferably, the preset neural network comprises a gaussian low-pass filtering algorithm;
and/or the number of the groups of groups,
the preset neural network comprises a convolutional neural network, and the convolutional neural network corresponds to a plurality of convolutional kernels which are configured randomly or fixedly in advance.
Preferably, the first image parameter includes an image size and/or an image resolution in a set direction;
the step of calculating the first pixel offset of the image pixel corresponding to the motion artifact in the first image based on the image parameter includes:
collecting a first displacement generated by a patient in the set direction when the patient moves in the corresponding examination scene;
calculating to obtain the first pixel offset corresponding to the image pixels in the first image in the set direction based on the first displacement and the image parameters;
and/or the number of the groups of groups,
the step of performing artifact correction processing on the image to be corrected by adopting a matched preset artifact correction network according to the target inspection scene and the offset parameter comprises the following steps:
acquiring an actual movement amount of the current patient in a set direction when the current patient moves in the target examination scene;
acquiring target image parameters of the image to be corrected;
wherein the target image parameters include an image size and/or an image resolution in the set direction;
and matching to obtain the corresponding preset artifact correction network based on the actual movement amount, the target image parameter and the target inspection scene, and carrying out artifact correction processing on the image to be corrected by adopting the preset artifact correction network.
Preferably, the correction method further includes:
constructing an image database based on a plurality of the first images;
the step of determining the target inspection scene corresponding to the image to be corrected comprises the following steps:
screening out second images with the matching degree with the images to be corrected being larger than a set threshold value from a plurality of first images of the image database;
and taking the inspection scene of the second image as the target inspection scene corresponding to the image to be corrected.
Preferably, the examination scene information corresponding to each examination scene includes at least one of patient basic information, patient posture information, examination system information, examination protocol information, and examination operation mode information.
Preferably, before the step of acquiring the image to be corrected of the current patient, the method further includes:
acquiring state change information of the current patient;
and if the state change information indicates that the current patient moves, executing the step of acquiring the image to be corrected of the current patient.
Preferably, the state change information includes at least one of positional deviation information, pressure change information, vibration change information, and optical signal change information generated when the patient's body moves.
The invention also provides a correction system for motion artifacts in images, the correction system comprising:
the image to be corrected acquisition module is used for acquiring an image to be corrected of the current patient;
the offset parameter acquisition module is used for acquiring offset parameters corresponding to motion artifacts generated by movement of the current patient in the target examination scene based on the image to be corrected;
and the artifact correction module is used for carrying out artifact correction processing on the image to be corrected by adopting a matched preset artifact correction network according to the target inspection scene and the offset parameter, and acquiring a target image of the current patient.
Preferably, the correction system further comprises:
the first image acquisition module is used for acquiring first images of different patients under different preset deflection angles under different types of examination scenes; wherein each first image is marked with an inspection scene;
the first image parameter acquisition module is used for acquiring first image parameters corresponding to the first image;
a first pixel offset calculating module, configured to calculate, based on the first image parameter, a first pixel offset of an image pixel corresponding to a motion artifact in the first image;
And the correction network construction module is used for training a preset neural network based on the first pixel offset and constructing a corresponding preset artifact correction network.
Preferably, the preset neural network comprises a gaussian low-pass filtering algorithm;
and/or the number of the groups of groups,
the preset neural network comprises a convolutional neural network, and the convolutional neural network corresponds to a plurality of convolutional kernels which are configured randomly or fixedly in advance.
Preferably, the first image parameter includes an image size and/or an image resolution in a set direction;
the first pixel offset amount calculation module includes:
collecting a first displacement generated by a patient in the set direction when the patient moves in the corresponding examination scene;
calculating to obtain the first pixel offset corresponding to the image pixels in the first image in the set direction based on the first displacement and the image parameters;
and/or the number of the groups of groups,
the artifact correction module includes:
an actual movement amount acquisition unit configured to acquire an actual movement amount in a set direction when the current patient moves in the target examination scene;
a target image parameter obtaining unit, configured to obtain a target image parameter of the image to be corrected;
Wherein the target image parameters include an image size and/or an image resolution in the set direction;
the artifact correction unit is used for obtaining a corresponding preset artifact correction network based on the actual movement amount, the target image parameters and the target inspection scene in a matching mode, and carrying out artifact correction processing on the image to be corrected by adopting the preset artifact correction network.
Preferably, the correction system further comprises:
an image database construction module for constructing an image database based on a plurality of the first images;
the image screening module is used for screening second images, the matching degree of which with the images to be corrected is larger than a set threshold, from a plurality of first images of the image database;
and the target inspection scene determining module is used for taking the inspection scene of the second image as the target inspection scene corresponding to the image to be corrected.
Preferably, the examination scene information corresponding to each examination scene includes at least one of patient basic information, patient posture information, examination system information, examination protocol information, and examination operation mode information.
Preferably, the correction system comprises:
The state change information acquisition module is used for acquiring the state change information of the current patient;
and the judging module is used for calling the image acquisition module to be corrected if the state change information characterizes the current patient to move.
Preferably, the state change information includes at least one of positional deviation information, pressure change information, vibration change information, and optical signal change information generated when the patient's body moves.
The invention also provides an electronic device comprising a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor realizes the correction method of the motion artifact in the image when executing the computer program.
The present invention also provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the above described method of correcting motion artifacts in images.
On the basis of conforming to the common knowledge in the field, the preferred conditions can be arbitrarily combined to obtain the preferred embodiments of the invention.
The invention has the positive progress effects that:
according to the invention, for any image to be corrected, the image to be corrected is an image containing motion artifacts generated by moving a current patient in a target inspection scene, the image pixel offset corresponding to the motion artifacts in the target inspection scene and the detected actual moving amount of the patient are calculated, a matched preset artifact correction network is adopted to perform artifact correction processing on the image to be corrected, the corrected target image of the current patient is finally transmitted to a scanning device for image display, so that the image containing the motion artifacts is optimized, the motion artifacts in the image are eliminated, the effect of displaying the image without the motion artifacts is finally displayed, the image quality is effectively improved, and the accuracy of patient illness state determination by subsequent doctors and other personnel is facilitated.
Drawings
Fig. 1 is a flowchart of a method for correcting motion artifacts in an image according to embodiment 1 of the present invention.
Fig. 2 is a first flowchart of a method for correcting motion artifacts in an image according to embodiment 2 of the present invention.
Fig. 3 is a second flowchart of the method for correcting motion artifact in an image according to embodiment 2 of the present invention.
Fig. 4 is a third flowchart of the method for correcting motion artifact in an image according to embodiment 2 of the present invention.
Fig. 5 is a block diagram of a system for correcting motion artifacts in images according to embodiment 3 of the present invention.
Fig. 6 is a block diagram of a system for correcting motion artifacts in images according to embodiment 4 of the present invention.
Fig. 7 is a schematic structural diagram of an electronic device according to embodiment 5 of the present invention.
Detailed Description
The invention is further illustrated by means of the following examples, which are not intended to limit the scope of the invention.
Example 1
As shown in fig. 1, the method for correcting motion artifacts in an image of the present embodiment includes:
s101, acquiring an image to be corrected of a current patient;
among them, the images to be corrected include, but are not limited to, nuclear magnetic resonance (Magnetic Resonance, MR) images, electronic computed tomography (Computed Tomography, CT) images, positron emission tomography (Positron Emission Computed Tomography, PET) images, ultrasound images, and the like. During the detection process, motion artifacts may be generated in the medical image due to autonomous or non-autonomous motion of the detection object. Involuntary movements (physiological movements) can produce physiological movement artifacts, which are mainly caused by respiratory movements, blood flow, systolic movements, etc. of the subject. Voluntary movements are mainly coughing, swallowing movements of the patient, body movements of the subject, etc. Taking CT images as an example, detecting motion of an object during a CT scan can disrupt the consistency and integrity of projection data. As another example, the pulsation of blood vessels, heart or the flow of cerebrospinal fluid during an MR scan can create artifacts of periodic motion in the phase encoding direction; autonomous patient movement may then result in parallel banding artifacts in the phase encoding direction.
S102, acquiring offset parameters corresponding to motion artifacts generated by movement of a current patient in a target examination scene based on an image to be corrected;
s103, performing artifact correction processing on the image to be corrected by adopting a matched preset artifact correction network according to the target inspection scene and the offset parameter, and obtaining a target image of the current patient.
Wherein the target image contains relatively little or no artifact information relative to the image to be corrected.
In this embodiment, for any image to be corrected, the image to be corrected is an image including motion artifacts generated by movement of a current patient in a target inspection scene, by calculating offset parameters corresponding to the motion artifacts in the target inspection scene, an artifact correction process is performed on the image to be corrected by using a matched preset artifact correction network, so as to obtain a target image of the current patient after correction, and finally, the target image is transmitted to a scanning device for image display, thereby optimizing the image including the motion artifacts, eliminating the motion artifacts therein, and finally displaying the image without the motion artifacts.
Example 2
The correction method of motion artifact in the image of the present embodiment is a further improvement of embodiment 1, specifically:
in an embodiment, as shown in fig. 2, step S101 further includes:
s1001, acquiring first images of different patients under different preset deflection angles under different types of examination scenes; wherein each first image is marked with an inspection scene;
s1002, acquiring a first image parameter corresponding to a first image;
s1003, calculating to obtain a first pixel offset of an image pixel corresponding to the motion artifact in the first image based on the first image parameter;
s1004, training a preset neural network based on the first pixel offset, and constructing a corresponding preset artifact correction network.
The preset artifact correction network corresponding to each different inspection scene is obtained through training, so that the real-time model is not required to be trained in a real-time manner, the total time length of the real-time model for artifact correction is saved, and the overall efficiency and rationality of image correction processing are effectively improved.
Wherein the same preset artifact correction network corresponds to one inspection scene;
specifically, presetting the neural network comprises processing an image by using a Gaussian low-pass filtering algorithm;
The process of training the neural network by using the Gaussian low-pass filtering method is as follows:
basic principle of gaussian low-pass filtering algorithm: acquiring Gaussian kernels, namely acquiring the weight of each pixel point around a central pixel value according to Gaussian distribution, and carrying out normalization processing; then, gaussian filtering calculation is carried out, a weighted average value of a pixel neighborhood is obtained to replace the pixel value of the point, and the weight of each neighborhood pixel point is monotonically increased along with the distance between the point and the center point; carrying out weighted average calculation on the current image on the basis of Gaussian kernel, namely carrying out convolution on the central pixel point, and carrying out artifact correction on the obtained image, namely the image after Gaussian blur, namely by smoothing the detail part of the image; the method comprises the steps of filtering high-frequency components of non-smooth parts in an image, retaining low-frequency components of smooth parts in the image, carrying out Gaussian blur on the image, and carrying out correlated suppression on Gaussian normal distribution noise, wherein the image becomes blurred, namely motion artifact in the image is corrected.
Before image processing, an image database indexed by image pixel offsets needs to be built. In this embodiment, first, a pixel offset is calculated according to the image size and the image resolution, and an image without motion artifact is obtained when the pixel offset is 0; then, images containing motion artifacts corresponding to different pixel offsets under different deflection angles including 0 °, 45 °, 90 °, 135 °, 180 °, 225 °, 270 °, 315 ° and the like are set, so that various scenes are constructed.
Further, the pixels of the image are determined according to the image size and the image resolution, when the monitoring device detects the displacement offset by the patient movement, the system obtains the offset of the pixels of the image corresponding to different motion offsets according to the motion offset and the pixel offset, so as to establish a motion artifact correction database (namely an image pixel offset database).
In one embodiment, the gaussian low pass filtering uses a two-dimensional gaussian filter formula, letting the center point coordinates be (0, 0):
where G (x, y) is a two-dimensional Gaussian function, σ is the variance, and x, y are the coordinate values of the point.
In another embodiment, two-dimensional gaussian filtering can be performed in two independent one-dimensional spaces, and the two-dimensional gaussian filtering is calculated separately, namely, the gaussian filtering calculation is performed in the x direction, and then the gaussian filtering calculation in the y direction is performed on the basis of the calculation:
through a Gaussian filtering method, the established motion artifact correction database is trained by a neural network, so that a corresponding preset artifact correction network can be obtained under each inspection scene, any image to be corrected can be matched with the preset artifact correction network to be corrected, the correction processing is performed, and the timeliness and the accuracy of the image to be corrected are ensured. The preset neural network comprises a convolutional neural network, and the convolutional neural network corresponds to a plurality of convolutional kernels which are configured randomly or fixedly in advance.
Of course, the preset training network can also adopt other model training algorithms, as long as the corresponding image artifact correction function can be realized as well, and the model training algorithm can be determined or adjusted according to actual requirements.
In an embodiment, the first image parameter comprises an image size and/or an image resolution in the set direction.
As shown in fig. 3, step S1003 specifically includes:
s10031, a step of calculating, based on an image parameter, a first pixel offset of an image pixel corresponding to a motion artifact in a first image, including:
s10032, collecting a first displacement generated in a set direction when a patient moves in a corresponding examination scene;
s10033, calculating to obtain a first pixel offset corresponding to an image pixel in the first image in the set direction based on the first displacement and the image parameter;
specifically, the calculation formula corresponding to step S10033 is as follows:
wherein Δpixel is the pixel offset in the x-direction, x is the image size in the x-direction, R x R x For the image resolution in the x-direction, Δx is the patient movement in the x-direction by the corresponding displacement offset.
In addition, in order to better train the images and obtain a more accurate neural network for removing motion artifacts, the images to be corrected including the motion artifacts are processed more accurately and efficiently, and a large number of motion images need to be acquired in various scanning scenes, for example, related parameters of the examination scenes such as images of different machine configurations, different patient sexes, different patient ages, different patient weights, different patient positions, different scanning protocols and the like need to be acquired, so that various clinical scenes are covered as far as possible.
In one embodiment, step S103 specifically includes:
s1031, acquiring the actual movement amount of the current patient in the set direction when moving in the target examination scene;
s1032, acquiring target image parameters of the image to be corrected;
wherein the target image parameters include image size and/or image resolution in a set direction;
s1033, based on the actual movement amount, the target image parameters and the target inspection scene, matching to obtain a corresponding preset artifact correction network, and carrying out artifact correction processing on the image to be corrected by adopting the preset artifact correction network.
In one embodiment, the correction method further comprises:
constructing an image database based on the plurality of first images;
the step of determining the target inspection scene corresponding to the image to be corrected comprises the following steps:
screening out second images with the matching degree with the images to be corrected being greater than a set threshold value from a plurality of first images in an image database;
screening out a first image with highest similarity as a second image matched with the image to be corrected; when a plurality of similarity is the same, selecting one first image at will as a second image matched with the image to be corrected, or generating reminding information and pushing the reminding information to related staff, and selecting a corresponding first image as the second image matched with the image to be corrected based on feedback information of the staff; the method is specifically adopted to determine the second image matched with the image to be corrected, and selection or setting determination can be performed according to actual scene requirements so as to more flexibly meet the control requirements of different actual scenes.
And taking the inspection scene of the second image as a target inspection scene corresponding to the image to be corrected.
Wherein, the examination scene information corresponding to each examination scene comprises patient basic information, patient posture information, examination system information, examination protocol information, examination operation mode information and the like.
Specifically, DICOM (digital imaging and communication in medicine) images of patients under different examination scenes are acquired, reconstructed images of patients with different body types under various different scanning scenes (such as different positioning, different movement distances and the like) are acquired as much as possible, and an image database under different examination scenes is established. Storing a raw data (Rawdata) image and a DICOM image on a preset server, and simultaneously distinguishing and storing images of different inspection scenes to the preset server by adopting a specific identification method; the classification schemes corresponding to different inspection scenes are shown in the following table 1:
TABLE 1
In the above table, HFS means head in front of, supine; HFP means head forward, prone; FFS means advanced foot, supine; FFP indicates advanced foot, prone; HFDL indicates head advanced, recumbent position. GRE represents a gradient echo pulse sequence; EPI represents a planar echo imaging sequence; FSE represents a fast spin echo sequence.
According to the classification scheme of the table, reconstructed images of various protocols of different patients under different examination scenes are collected, and because the main purpose of the scheme is to correct motion artifacts, the reconstructed images of the patients under the scenes without position deviation, deviation of 1cm, deviation of 2cm, deviation of 3cm … … and the like are required to be collected, various deflection angles (such as 0, 45, 90, 135, 180, 225, 270 and 315) are considered, various movement modes which can occur to the patients during actual examination are simulated as much as possible, the images are stored on a specific server, and are managed in different categories, and an image database (or a motion correction database) for multi-scene analysis of the reconstructed images is created.
Through the fine construction of the image database, the images of any patient in any moving state can be obtained by matching and inquiring in the constructed image database in an actual examination scene, and the realizability of the subsequent artifact correction operation and the accuracy of matching correction can be further ensured with high quality.
Of course, other classification criteria than the above may be used to acquire reconstructed images of the patient under more examination scenes, and specifically may be flexibly determined or adjusted according to the actual situation.
In an embodiment, as shown in fig. 4, step S101 further includes:
s1005, acquiring state change information of a current patient;
s1006, if the state change information indicates that the current patient moves, step S101 is executed.
The state change information includes positional deviation information, pressure change information, vibration change information, optical signal change information, and the like, which are generated when the patient body moves.
For example, a 3D camera is arranged in a detection device (such as a nuclear magnetic resonance device) to acquire images of a patient, a pressure sensor is arranged to detect pressure value changes within a set time period, a vibration sensor is arranged to detect vibration value changes within the set time period, a light sensor is arranged to detect light signal changes within the set time period, and the like, and corresponding position movement conditions are analyzed and determined in time.
In addition, two or more types of sensors can be simultaneously arranged in the detection equipment, false triggering can be effectively avoided, so that a more timely and accurate triggering control effect is achieved, the artifact correction process is prevented from being executed in an unnecessary scene, the waste of computing resources is avoided, the image generation process is caused to be slower, the actual requirements are not met, and the like, so that the accuracy, the rationality and the timeliness of the image artifact correction control are improved.
The implementation principle of the method for correcting motion artifacts in an image based on the deep learning algorithm in this embodiment is specifically described below with reference to an example and fig. 4:
(1) Building an image database
According to a set classification scheme, collecting reconstructed images of various protocols of different patients under different examination scenes, collecting the reconstructed images of the patients under different deflection scenes, considering various deflection angles and the like, simulating various movement modes possibly occurring to the patients during actual examination as much as possible, storing the images on a specific server, managing the images according to categories, and creating an image database for multi-scene analysis of the reconstructed images;
(2) Whether or not artifact correction triggers
Acquiring state change information of a patient, including but not limited to position deviation, pressure change, vibration change, optical signal change and the like, timely and accurately judging whether the current patient moves in the examination process, judging that the scanned image positively contains motion artifact once the current patient moves, and automatically triggering an artifact correction function; otherwise, the image is directly reconstructed without triggering.
(3) Presetting artifact correction network
Acquiring first images of different patients under different preset deflection angles under different types of examination scenes, wherein each first image is marked with the examination scene; acquiring a first image parameter (size, resolution and the like) corresponding to the first image, and calculating a pixel displacement of a corresponding image pixel according to a displacement generated by the detected patient movement; and training a preset neural network based on the pixel offset by adopting a Gaussian low-pass filtering method, and constructing a corresponding preset artifact correction network.
The model training is performed on an image level, but not on the K space data, so that the efficiency and the accuracy of the model training are effectively ensured.
In addition, the preset artifact correction network corresponding to each different inspection scene can be obtained through pre-training before the actual inspection, so that the actual inspection scene can be directly matched and called without training a model in real time;
the real-time model training or the pre-model training is adopted, and the real-time model training or the pre-model training can be selected or adjusted according to actual requirements.
(4) Image artifact correction
Comprehensively determining a preset artifact correction network matched with the to-be-corrected image based on an inspection scene corresponding to the to-be-corrected image, the pixel offset of the image pixels, the image size and the image resolution; inputting an image to be corrected into a preset artifact correction network, directly carrying out artifact correction processing on the image to be corrected through the preset artifact correction network, optimizing the image containing motion artifacts at the image level, eliminating the motion artifacts therein, and displaying the image without the motion artifacts to obtain the DICOM scanning image after the motion artifacts are removed. In addition, the image artifact correction in the embodiment can support on-line correction and off-line correction, and can more flexibly meet more required image artifact correction scenes.
After scanning by adopting scanning equipment to obtain an image to be corrected containing motion artifact, calculating according to the motion offset generated when a monitoring device acquires the current patient, calculating image parameters such as the image size, the image resolution and the like of the image to be corrected to obtain a corresponding pixel offset, and further combining the current examination such as the current machine configuration, the patient age, the patient sex, the patient weight, the patient positioning, the patient scanning sequence and the like, and searching and obtaining a preset artifact correction network which is most in line with the current artifact image in the current examination scene in a database according to the information; and finally, outputting the image subjected to motion artifact correction by a preset artifact correction network, and transmitting the image to a scanning device for image display.
In this embodiment, for any image to be corrected, the image to be corrected is an image including motion artifacts generated by moving a current patient in a target inspection scene, by calculating an image pixel offset corresponding to the motion artifacts in the target inspection scene and an actual moving amount of the detected patient, an artifact correction process is performed on the image to be corrected by using a matched preset artifact correction network, so as to obtain a target image of the corrected current patient, and finally, the target image is transmitted to a scanning device for image display, thereby optimizing an image including the motion artifacts, eliminating the motion artifacts therein, and finally displaying an image without the motion artifacts.
Example 3
As shown in fig. 5, the correction system of motion artifact in the image of the present embodiment includes:
the image to be corrected acquisition module 1 is used for acquiring an image to be corrected of a current patient;
the image to be corrected includes, but is not limited to, nuclear magnetic resonance images, electron computer tomography images, positron emission type computer tomography images, ultrasound images, and the like. During the detection process, motion artifacts may be generated in the medical image due to autonomous or non-autonomous motion of the detection object. Involuntary movements (physiological movements) can produce physiological movement artifacts, which are mainly caused by respiratory movements, blood flow, systolic movements, etc. of the subject. Voluntary movements are mainly coughing, swallowing movements of the patient, body movements of the subject, etc. Taking CT images as an example, detecting motion of an object during a CT scan can disrupt the consistency and integrity of projection data. As another example, the pulsation of blood vessels, heart or the flow of cerebrospinal fluid during an MR scan can create artifacts of periodic motion in the phase encoding direction; autonomous patient movement may then result in parallel banding artifacts in the phase encoding direction.
The offset parameter acquisition module 2 is used for acquiring offset parameters corresponding to motion artifacts generated by movement of a current patient in a target examination scene based on an image to be corrected;
And the artifact correction module 3 is used for carrying out artifact correction processing on the image to be corrected by adopting a matched preset artifact correction network according to the target inspection scene and the offset parameter, and obtaining the target image of the current patient.
Wherein the target image contains relatively little or no artifact information relative to the image to be corrected.
In this embodiment, for any image to be corrected, the image to be corrected is an image including motion artifacts generated by movement of a current patient in a target inspection scene, by calculating offset parameters corresponding to the motion artifacts in the target inspection scene, an artifact correction process is performed on the image to be corrected by using a matched preset artifact correction network, so as to obtain a target image of the current patient after correction, and finally, the target image is transmitted to a scanning device for image display, thereby optimizing the image including the motion artifacts, eliminating the motion artifacts therein, and finally displaying the image without the motion artifacts.
Example 4
As shown in fig. 6, the correction system of motion artifact in the image of the present embodiment is a further improvement of embodiment 3, specifically:
In one embodiment, the correction system further comprises:
the first image acquisition module 4 is used for acquiring first images of different patients under different preset deflection angles under different types of examination scenes; wherein each first image is marked with an inspection scene;
the first image parameter obtaining module 5 is used for obtaining first image parameters corresponding to the first image;
a first pixel offset calculating module 6, configured to calculate, based on the first image parameter, a first pixel offset of an image pixel corresponding to the motion artifact in the first image;
the correction network construction module 7 is configured to train the preset neural network based on the first pixel offset, and construct a corresponding preset artifact correction network.
In one embodiment, the presetting the neural network includes processing the image using a gaussian low pass filtering algorithm;
the process of training the neural network by using the Gaussian low-pass filtering method is as follows:
basic principle of gaussian low-pass filtering algorithm: acquiring Gaussian kernels, namely acquiring the weight of each pixel point around a central pixel value according to Gaussian distribution, and carrying out normalization processing; then, gaussian filtering calculation is carried out, a weighted average value of a pixel neighborhood is obtained to replace the pixel value of the point, and the weight of each neighborhood pixel point is monotonically increased along with the distance between the point and the center point; carrying out weighted average calculation on the current image on the basis of Gaussian kernel, namely carrying out convolution on the central pixel point, and carrying out artifact correction on the obtained image, namely the image after Gaussian blur, namely by smoothing the detail part of the image; the method comprises the steps of filtering high-frequency components of non-smooth parts in an image, retaining low-frequency components of smooth parts in the image, carrying out Gaussian blur on the image, and carrying out correlated suppression on Gaussian normal distribution noise, wherein the image becomes blurred, namely motion artifact in the image is corrected.
Before image processing, an image database indexed by image pixel offsets needs to be built. In this embodiment, first, a pixel offset is calculated according to the image size and the image resolution, and an image without motion artifact is obtained when the pixel offset is 0; then, images containing motion artifacts corresponding to different pixel offsets under different deflection angles including 0 °, 45 °, 90 °, 135 °, 180 °, 225 °, 270 °, 315 ° and the like are set, so that various scenes are constructed.
Further, the pixels of the image are determined according to the image size and the image resolution, when the monitoring device detects the displacement offset by the patient movement, the system obtains the offset of the pixels of the image corresponding to different motion offsets according to the motion offset and the pixel offset, so as to establish a motion artifact correction database (namely an image pixel offset database).
In one embodiment, the gaussian low pass filtering uses a two-dimensional gaussian filter formula, letting the center point coordinates be (0, 0):
where G (x, y) is a two-dimensional Gaussian function, σ is the variance, and x, y are the coordinate values of the point.
In another embodiment, two-dimensional gaussian filtering can be performed in two independent one-dimensional spaces, and the two-dimensional gaussian filtering is calculated separately, namely, the gaussian filtering calculation is performed in the x direction, and then the gaussian filtering calculation in the y direction is performed on the basis of the calculation:
Through a Gaussian filtering method, the established motion artifact correction database is trained by a neural network, so that a corresponding preset artifact correction network can be obtained under each inspection scene, any image to be corrected can be matched with the preset artifact correction network to be corrected, the correction processing is performed, and the timeliness and the accuracy of the image to be corrected are ensured.
The preset neural network further comprises a convolutional neural network, and the convolutional neural network corresponds to a plurality of convolutional kernels which are configured randomly or fixedly in advance.
In an embodiment, the first image parameter comprises an image size and/or an image resolution in a set direction;
the first pixel shift amount calculation module 6 includes:
a first displacement amount acquisition unit 8 for acquiring a first displacement amount generated in a set direction when a patient moves in a corresponding examination scene;
a first pixel offset calculating unit 9, configured to calculate, based on the first displacement and the image parameter, a first pixel offset corresponding to an image pixel in the first image in the set direction;
specifically, the calculation formula corresponding to the first pixel offset calculated by the first pixel offset calculation unit is as follows:
Wherein Δpixel is the pixel offset in the x-direction, x is the image size in the x-direction, R x R x For the image resolution in the x-direction, Δx is the patient movement in the x-direction by the corresponding displacement offset.
In addition, in order to better train the images and obtain a more accurate neural network for removing motion artifacts, the images to be corrected including the motion artifacts are processed more accurately and efficiently, and a large number of motion images need to be acquired in various scanning scenes, for example, related parameters of the examination scenes such as images of different machine configurations, different patient sexes, different patient ages, different patient weights, different patient positions, different scanning protocols and the like need to be acquired, so that various clinical scenes are covered as far as possible.
In one embodiment, the artifact correction module 3 comprises:
an actual movement amount acquisition unit 10 for acquiring an actual movement amount in a set direction when a current patient moves in a target examination scene;
a target image parameter acquiring unit 11 for acquiring target image parameters of an image to be corrected;
wherein the target image parameters include image size and/or image resolution in a set direction;
the artifact correction unit 12 is configured to match the actual movement amount, the target image parameter, and the target inspection scene to obtain a corresponding preset artifact correction network, and perform artifact correction processing on the image to be corrected by using the preset artifact correction network.
In one embodiment, the correction system further comprises:
an image database construction module 13 for constructing an image database based on the plurality of first images;
the image screening module 14 is used for screening second images with the matching degree with the images to be corrected being larger than a set threshold value from a plurality of first images in the image database;
screening out a first image with highest similarity as a second image matched with the image to be corrected; when a plurality of similarity is the same, selecting one first image at will as a second image matched with the image to be corrected, or generating reminding information and pushing the reminding information to related staff, and selecting a corresponding first image as the second image matched with the image to be corrected based on feedback information of the staff; the method is specifically adopted to determine the second image matched with the image to be corrected, and selection or setting determination can be performed according to actual scene requirements so as to more flexibly meet the control requirements of different actual scenes.
The target inspection scene determining module 15 is configured to take the inspection scene of the second image as a target inspection scene corresponding to the image to be corrected.
In an embodiment, the examination scene information corresponding to each examination scene includes at least one of patient basic information, patient posture information, examination system information, examination protocol information, examination operation mode information.
Specifically, DICOM (digital imaging and communication in medicine) images of patients under different examination scenes are acquired, reconstructed images of patients with different body types under various different scanning scenes (such as different positioning, different movement distances and the like) are acquired as much as possible, and an image database under different examination scenes is established. Storing a raw data (Rawdata) image and a DICOM image on a preset server, and simultaneously distinguishing and storing images of different inspection scenes to the preset server by adopting a specific identification method; the classification schemes corresponding to different inspection scenes are shown in the following table 1:
TABLE 1
In the above table, HFS means head in front of, supine; HFP means head forward, prone; FFS means advanced foot, supine; FFP indicates advanced foot, prone; HFDL indicates head advanced, recumbent position. GRE represents a gradient echo pulse sequence; EPI represents a planar echo imaging sequence; FSE represents a fast spin echo sequence.
According to the classification scheme of the table, reconstructed images of various protocols of different patients under different examination scenes are collected, and because the main purpose of the scheme is to correct motion artifacts, the reconstructed images of the patients under the scenes without position deviation, deviation of 1cm, deviation of 2cm, deviation of 3cm … … and the like are required to be collected, various deflection angles (such as 0, 45, 90, 135, 180, 225, 270 and 315) are considered, various movement modes which can occur to the patients during actual examination are simulated as much as possible, the images are stored on a specific server, and are managed in different categories, and an image database (or a motion correction database) for multi-scene analysis of the reconstructed images is created.
Through the fine construction of the image database, the images of any patient in any moving state can be obtained by matching and inquiring in the constructed image database in an actual examination scene, and the realizability of the subsequent artifact correction operation and the accuracy of matching correction can be further ensured with high quality.
Of course, other classification criteria than the above may be used to acquire reconstructed images of the patient under more examination scenes, and specifically may be flexibly determined or adjusted according to the actual situation.
In one embodiment, the correction system includes:
a state change information acquisition module 16 for acquiring state change information of the current patient;
the judging module 17 is configured to invoke the image obtaining module 1 to be corrected if the state change information indicates that the current patient moves.
In one embodiment, the state change information includes positional shift information, pressure change information, vibration change information, light signal change information, and the like generated when the patient's body moves.
For example, a 3D camera is arranged in a detection device (such as a nuclear magnetic resonance device) to acquire images of a patient, a pressure sensor is arranged to detect pressure value changes within a set time period, a vibration sensor is arranged to detect vibration value changes within the set time period, a light sensor is arranged to detect light signal changes within the set time period, and the like, and corresponding position movement conditions are analyzed and determined in time.
In addition, two or more types of sensors can be simultaneously arranged in the detection equipment, false triggering can be effectively avoided, so that a more timely and accurate triggering control effect is achieved, the artifact correction process is prevented from being executed in an unnecessary scene, the waste of computing resources is avoided, the image generation process is caused to be slower, the actual requirements are not met, and the like, so that the accuracy, the rationality and the timeliness of the image artifact correction control are improved.
The implementation principle of the method for correcting motion artifacts in an image based on the deep learning algorithm in this embodiment is specifically described below with reference to an example and fig. 4:
(1) Building an image database
According to a set classification scheme, collecting reconstructed images of various protocols of different patients under different examination scenes, collecting the reconstructed images of the patients under different deflection scenes, considering various deflection angles and the like, simulating various movement modes possibly occurring to the patients during actual examination as much as possible, storing the images on a specific server, managing the images according to categories, and creating an image database for multi-scene analysis of the reconstructed images;
(2) Whether or not artifact correction triggers
Acquiring state change information of a patient, including but not limited to position deviation, pressure change, vibration change, optical signal change and the like, timely and accurately judging whether the current patient moves in the examination process, judging that the scanned image positively contains motion artifact once the current patient moves, and automatically triggering an artifact correction function; otherwise, the image is directly reconstructed without triggering.
(3) Presetting artifact correction network
Acquiring first images of different patients under different preset deflection angles under different types of examination scenes, wherein each first image is marked with the examination scene; acquiring a first image parameter (size, resolution and the like) corresponding to the first image, and calculating a pixel displacement of a corresponding image pixel according to a displacement generated by the detected patient movement; and training a preset neural network based on the pixel offset by adopting a Gaussian low-pass filtering method, and constructing a corresponding preset artifact correction network.
The model training is performed on an image level, but not on the K space data, so that the efficiency and the accuracy of the model training are effectively ensured.
In addition, the preset artifact correction network corresponding to each different inspection scene can be obtained through pre-training before the actual inspection, so that the actual inspection scene can be directly matched and called without training a model in real time;
The real-time model training or the pre-model training is adopted, and the real-time model training or the pre-model training can be selected or adjusted according to actual requirements.
(4) Image artifact correction
Comprehensively determining a preset artifact correction network matched with the to-be-corrected image based on an inspection scene corresponding to the to-be-corrected image, the pixel offset of the image pixels, the image size and the image resolution; inputting an image to be corrected into a preset artifact correction network, directly carrying out artifact correction processing on the image to be corrected through the preset artifact correction network, optimizing the image containing motion artifacts at the image level, eliminating the motion artifacts therein, and displaying the image without the motion artifacts to obtain the DICOM scanning image after the motion artifacts are removed. In addition, the image artifact correction in the embodiment can support on-line correction and off-line correction, and can more flexibly meet more required image artifact correction scenes.
After scanning by adopting scanning equipment to obtain an image to be corrected containing motion artifact, calculating according to the motion offset generated when a monitoring device acquires the current patient, calculating image parameters such as the image size, the image resolution and the like of the image to be corrected to obtain a corresponding pixel offset, and further combining the current examination such as the current machine configuration, the patient age, the patient sex, the patient weight, the patient positioning, the patient scanning sequence and the like, and searching and obtaining a preset artifact correction network which is most in line with the current artifact image in the current examination scene in a database according to the information; and finally, outputting the image subjected to motion artifact correction by a preset artifact correction network, and transmitting the image to a scanning device for image display.
In this embodiment, for any image to be corrected, the image to be corrected is an image including motion artifacts generated by moving a current patient in a target inspection scene, by calculating an image pixel offset corresponding to the motion artifacts in the target inspection scene and an actual moving amount of the detected patient, an artifact correction process is performed on the image to be corrected by using a matched preset artifact correction network, so as to obtain a target image of the corrected current patient, and finally, the target image is transmitted to a scanning device for image display, thereby optimizing an image including the motion artifacts, eliminating the motion artifacts therein, and finally displaying an image without the motion artifacts.
Example 5
Fig. 7 is a schematic structural diagram of an electronic device according to embodiment 5 of the present invention. The electronic device comprises a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the methods of the above embodiments when executing the program. The electronic device 30 shown in fig. 7 is only an example and should not be construed as limiting the functionality and scope of use of embodiments of the present invention.
As shown in fig. 7, the electronic device 30 may be in the form of a general purpose computing device, which may be a server device, for example. Components of electronic device 30 may include, but are not limited to: the at least one processor 31, the at least one memory 32, a bus 33 connecting the different system components, including the memory 32 and the processor 31.
The bus 33 includes a data bus, an address bus, and a control bus.
Memory 32 may include volatile memory such as Random Access Memory (RAM) 321 and/or cache memory 322, and may further include Read Only Memory (ROM) 323.
Memory 32 may also include a program/utility 325 having a set (at least one) of program modules 324, such program modules 324 including, but not limited to: an operating system, one or more application programs, other program modules, and program data, each or some combination of which may include an implementation of a network environment.
The processor 31 executes various functional applications and data processing, such as the method of the above-described embodiment of the present invention, by running a computer program stored in the memory 32.
The electronic device 30 may also communicate with one or more external devices 34 (e.g., keyboard, pointing device, etc.). Such communication may be through an input/output (I/O) interface 35. Also, model-generating device 30 may also communicate with one or more networks, such as a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public network, such as the internet, via network adapter 36. As shown in fig. 7, network adapter 36 communicates with the other modules of model-generating device 30 via bus 33. It should be appreciated that although not shown in the figures, other hardware and/or software modules may be used in connection with the model-generating device 30, including, but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID (disk array) systems, tape drives, data backup storage systems, and the like.
It should be noted that although several units/modules or sub-units/modules of an electronic device are mentioned in the above detailed description, such a division is merely exemplary and not mandatory. Indeed, the features and functionality of two or more units/modules described above may be embodied in one unit/module in accordance with embodiments of the present invention. Conversely, the features and functions of one unit/module described above may be further divided into ones that are embodied by a plurality of units/modules.
Example 6
The present embodiment provides a computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of the method of the above embodiment.
More specifically, among others, readable storage media may be employed including, but not limited to: portable disk, hard disk, random access memory, read only memory, erasable programmable read only memory, optical storage device, magnetic storage device, or any suitable combination of the foregoing.
In a possible implementation, the invention may also be implemented in the form of a program product comprising program code for causing a terminal device to carry out the steps of the method of implementing the embodiments described above, when the program product is run on the terminal device.
Wherein the program code for carrying out the invention may be written in any combination of one or more programming languages, the program code may be executed entirely on the user device, partially on the user device, as a stand-alone software package, partially on the user device, partially on a remote device or entirely on the remote device.
While specific embodiments of the invention have been described above, it will be appreciated by those skilled in the art that this is by way of example only, and the scope of the invention is defined by the appended claims. Various changes and modifications to these embodiments may be made by those skilled in the art without departing from the principles and spirit of the invention, but such changes and modifications fall within the scope of the invention.
Claims (10)
1. A method of correcting motion artifacts in an image, the method comprising:
acquiring an image to be corrected of a current patient;
acquiring offset parameters corresponding to motion artifacts generated by movement of the current patient in the target examination scene based on the image to be corrected;
and according to the target examination scene and the offset parameter, performing artifact correction processing on the image to be corrected by adopting a matched preset artifact correction network, and acquiring a target image of the current patient.
2. The method for correcting motion artifacts in images according to claim 1, characterized in that before the step of obtaining the image to be corrected of the current patient, it further comprises:
acquiring first images of different patients under different preset deflection angles under different types of examination scenes; wherein each first image is marked with an inspection scene;
acquiring a first image parameter corresponding to the first image;
calculating a first pixel offset of an image pixel corresponding to the motion artifact in the first image based on the first image parameter;
training a preset neural network based on the first pixel offset, and constructing a corresponding preset artifact correction network.
3. The method for correcting motion artifacts in images according to claim 2, characterized in that said preset neural network comprises a gaussian low pass filtering algorithm;
and/or the number of the groups of groups,
the preset neural network comprises a convolutional neural network, and the convolutional neural network corresponds to a plurality of convolutional kernels which are configured randomly or fixedly in advance.
4. A method of correcting motion artefacts in an image as claimed in claim 2, characterized in that the first image parameter comprises an image size and/or an image resolution in a set direction;
The step of calculating the first pixel offset of the image pixel corresponding to the motion artifact in the first image based on the image parameter includes:
collecting a first displacement generated by a patient in the set direction when the patient moves in the corresponding examination scene;
calculating to obtain the first pixel offset corresponding to the image pixels in the first image in the set direction based on the first displacement and the image parameters;
and/or the number of the groups of groups,
the step of performing artifact correction processing on the image to be corrected by adopting a matched preset artifact correction network according to the target inspection scene and the offset parameter comprises the following steps:
acquiring an actual movement amount of the current patient in a set direction when the current patient moves in the target examination scene;
acquiring target image parameters of the image to be corrected;
wherein the target image parameters include an image size and/or an image resolution in the set direction;
and matching to obtain the corresponding preset artifact correction network based on the actual movement amount, the target image parameter and the target inspection scene, and carrying out artifact correction processing on the image to be corrected by adopting the preset artifact correction network.
5. The method of correcting motion artifacts in images according to any one of claims 2 to 4, further comprising:
constructing an image database based on a plurality of the first images;
the step of determining the target inspection scene corresponding to the image to be corrected comprises the following steps:
screening out second images with the matching degree with the images to be corrected being larger than a set threshold value from a plurality of first images of the image database;
and taking the inspection scene of the second image as the target inspection scene corresponding to the image to be corrected.
6. The method of claim 5, wherein the examination scene information corresponding to each of the examination scenes includes at least one of patient basic information, patient posture information, examination system information, examination protocol information, and examination operation mode information.
7. The method for correcting motion artifacts in images according to any one of claims 1 to 4, characterized in that before the step of obtaining the image to be corrected of the current patient, it further comprises:
acquiring state change information of the current patient;
and if the state change information indicates that the current patient moves, executing the step of acquiring the image to be corrected of the current patient.
8. The method of claim 7, wherein the state change information includes at least one of positional shift information, pressure change information, vibration change information, and optical signal change information generated when the patient's body moves.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements a method of correcting motion artifacts in images according to any one of claims 1-8 when the computer program is executed.
10. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements a method of correcting motion artifacts in images according to any one of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211041644.6A CN117689582A (en) | 2022-08-29 | 2022-08-29 | Correction method for motion artifact in image, electronic device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211041644.6A CN117689582A (en) | 2022-08-29 | 2022-08-29 | Correction method for motion artifact in image, electronic device and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117689582A true CN117689582A (en) | 2024-03-12 |
Family
ID=90132460
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211041644.6A Pending CN117689582A (en) | 2022-08-29 | 2022-08-29 | Correction method for motion artifact in image, electronic device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117689582A (en) |
-
2022
- 2022-08-29 CN CN202211041644.6A patent/CN117689582A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110121749B (en) | Deep learning medical system and method for image acquisition | |
| CN110114834B (en) | Deep learning medical system and method for medical procedures | |
| US10074038B2 (en) | Deep learning medical systems and methods for image reconstruction and quality evaluation | |
| US10304198B2 (en) | Automatic medical image retrieval | |
| US10952694B2 (en) | Method and apparatus for correcting computed tomography image | |
| EP3355273B1 (en) | Coarse orientation detection in image data | |
| US20200058098A1 (en) | Image processing apparatus, image processing method, and image processing program | |
| CN110827956A (en) | System and method for identifying medical image acquisition parameters | |
| US9336457B2 (en) | Adaptive anatomical region prediction | |
| JP2022517769A (en) | 3D target detection and model training methods, equipment, equipment, storage media and computer programs | |
| US10867375B2 (en) | Forecasting images for image processing | |
| CN107252353B (en) | Control method of medical imaging equipment and medical imaging equipment | |
| EP3758602A1 (en) | Neural network classification of osteolysis and synovitis near metal implants | |
| CN111260647A (en) | CT scanning auxiliary method based on image detection, computer readable storage medium and CT scanning device | |
| JP6995535B2 (en) | Image processing equipment, image processing methods and programs | |
| US11837352B2 (en) | Body representations | |
| CN117689582A (en) | Correction method for motion artifact in image, electronic device and storage medium | |
| CN115761134A (en) | Twin model generation method, system, nuclear magnetic resonance device, and medium | |
| JP5051025B2 (en) | Image generating apparatus, program, and image generating method | |
| US20210192717A1 (en) | Systems and methods for identifying atheromatous plaques in medical images | |
| WO2024095434A1 (en) | Image processing program, image processing method, detection program, detection method, and information processing device | |
| US11836923B2 (en) | Image processing apparatus, image processing method, and storage medium | |
| WO2023103975A1 (en) | Medical image movement detection and correction method and system, and computer readable medium | |
| Jayadharshini et al. | Deep Learning Strategies in Biomedicine Imaging Technique | |
| WO2023183823A1 (en) | Deep learning-based medical image motion artifact correction |
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 |