CN115100310A - Automatic prompting method and system for magnetic resonance magnetic sensitivity artifact - Google Patents
Automatic prompting method and system for magnetic resonance magnetic sensitivity artifact Download PDFInfo
- Publication number
- CN115100310A CN115100310A CN202210735986.1A CN202210735986A CN115100310A CN 115100310 A CN115100310 A CN 115100310A CN 202210735986 A CN202210735986 A CN 202210735986A CN 115100310 A CN115100310 A CN 115100310A
- Authority
- CN
- China
- Prior art keywords
- magnetic
- artifact
- magnetic resonance
- resonance image
- field distribution
- 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
- 230000035945 sensitivity Effects 0.000 title claims abstract description 51
- 238000000034 method Methods 0.000 title claims abstract description 31
- 230000004044 response Effects 0.000 claims abstract description 7
- 230000003321 amplification Effects 0.000 claims description 15
- 238000003199 nucleic acid amplification method Methods 0.000 claims description 15
- 238000010586 diagram Methods 0.000 claims description 8
- 238000013136 deep learning model Methods 0.000 claims description 7
- 238000002372 labelling Methods 0.000 claims description 5
- 230000009977 dual effect Effects 0.000 claims description 4
- 238000013434 data augmentation Methods 0.000 claims 1
- 230000003902 lesion Effects 0.000 abstract description 12
- 210000001519 tissue Anatomy 0.000 description 8
- 238000002597 diffusion-weighted imaging Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 230000008859 change Effects 0.000 description 4
- 238000002595 magnetic resonance imaging Methods 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 3
- 238000003384 imaging method Methods 0.000 description 3
- 239000000126 substance Substances 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000002591 computed tomography Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 230000005855 radiation Effects 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 210000001154 skull base Anatomy 0.000 description 2
- 210000001015 abdomen Anatomy 0.000 description 1
- 238000010521 absorption reaction Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000002457 bidirectional effect Effects 0.000 description 1
- 210000000988 bone and bone Anatomy 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 210000003695 paranasal sinus Anatomy 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 239000000941 radioactive substance Substances 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/003—Reconstruction from projections, e.g. tomography
- G06T11/008—Specific post-processing after tomographic reconstruction, e.g. voxelisation, metal artifact correction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/90—Dynamic range modification of images or parts thereof
- G06T5/94—Dynamic range modification of images or parts thereof based on local image properties, e.g. for local contrast enhancement
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10088—Magnetic resonance imaging [MRI]
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Magnetic Resonance Imaging Apparatus (AREA)
Abstract
The invention provides an automatic prompting method and system for magnetic resonance magnetic sensitivity artifacts, belonging to the technical field of image processing and comprising the following steps: acquiring a magnetic resonance image and acquiring a corresponding magnetic field distribution map; inputting the magnetic resonance image and the magnetic field distribution map into an artifact prediction model to obtain whether the magnetic resonance image contains magnetic sensitivity artifacts; in response to the magnetic resonance image containing magnetic susceptibility artifacts, the locations of all the magnetic susceptibility artifacts are labeled and the probability of each magnetic susceptibility artifact is determined. Has the advantages that: the invention can predict whether the magnetic resonance image contains the magnetic sensitivity artifact or not through the model, automatically identify the magnetic sensitivity artifact and label and display the magnetic sensitivity artifact in response to the magnetic resonance image containing the magnetic sensitivity artifact, and distinguish the magnetic sensitivity artifact from possible real lesions, thereby effectively reducing the misdiagnosis problem of doctors caused by the artifact.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to an automatic prompting method and system for magnetic resonance magnetic sensitivity artifacts.
Background
Magnetic Resonance Imaging (MRI) is a new examination technique that is used based on the principle that nuclei with magnetic distances can make transitions between energy levels under the action of a magnetic field. MRI is realized by radiating energy to generate signals from substances in the body to the surrounding environment through the action of an external high-frequency magnetic field, the imaging process is similar to image reconstruction and Computed Tomography (CT), only MRI does not depend on radiation, absorption and reflection of the outside or gamma radiation of radioactive substances in the body, but images by utilizing the interaction of an external magnetic field and the objects, and the high-energy magnetic field is harmless to the human body.
Magnetic susceptibility artifacts (susceptibility artifacts) are one type of artifact that often occurs during magnetic resonance imaging, also known as susceptibility artifacts. The reason for this is that different substances have different magnetic susceptibilities, and when the magnetic susceptibilities of different substances are greatly different, the local magnetic field at the boundary will be non-uniform, resulting in the distortion and error of the local signal. Occur at interfaces of different magnetic susceptibility materials, typically at bone/air interfaces (including paranasal sinuses, skull base, sphenoid saddles, etc.), tissue/air interfaces, such as skull base, abdomen, etc. As shown in fig. 1, for Diffusion Weighted Imaging (DWI) acquired by Echo Planar Imaging (EPI), the magnetic sensitivity artifacts in the DWI image are particularly prominent, and mainly appear as a highlight signal with abnormal edges, which is easily confused with lesions, resulting in misdiagnosis.
In the prior art, a common method for alleviating the magnetic sensitivity artifact includes improving the quality of shimming, using a special acquisition sequence, such as multiple excitation acquisition (multishot acquisition), and performing artifact correction in a subsequent processing process. However, these processing methods cannot completely and completely remove the magnetic sensitivity artifacts, and the signal-to-noise ratio of the image is reduced by using a special acquisition sequence or an artifact correction process.
Disclosure of Invention
In order to solve the technical problems, the invention provides an automatic prompting method and system for magnetic resonance magnetic sensitivity artifacts.
The technical problem solved by the invention can be realized by adopting the following technical scheme:
a method for automatically prompting magnetic resonance magnetic sensitivity artifact comprises the following steps:
acquiring a magnetic resonance image and acquiring a corresponding magnetic field distribution map;
inputting the magnetic resonance image and the magnetic field distribution map into an artifact prediction model to obtain whether the magnetic resonance image contains magnetic sensitivity artifacts;
and responding to the magnetic resonance image containing magnetic susceptibility artifacts, labeling the positions of all the magnetic susceptibility artifacts, and determining the probability of each magnetic susceptibility artifact.
Preferably, acquiring a corresponding magnetic field profile comprises:
acquiring an echo image; and calculating the magnetic field distribution map according to the phases of the two echo images.
Preferably, the echo image comprises a dual echo gradient echo image or a bi-directional phase encoded planar echo image.
Preferably, the artifact prediction model is a deep learning model.
Preferably, the magnetic resonance image and the magnetic field distribution map are input to an artifact prediction model, comprising:
and splicing the magnetic resonance image and the magnetic field distribution map in a channel dimension according to a preset combination mode, and inputting the artifact prediction model layer by layer in a two-dimensional form.
Preferably, the preset combination mode includes any one of an absolute value, an absolute value and a phase, and a real part and an imaginary part of the magnetic resonance image.
Preferably, the artifact prediction model is obtained by pre-training the magnetic resonance image and the corresponding magnetic field distribution map as a training data set, and before training, the artifact prediction model further includes:
performing data amplification on the training data set.
Preferably, the data amplification comprises at least: the magnetic resonance image is locally brightened to simulate diseased tissue.
The invention also provides an automatic prompting system for magnetic resonance magnetic sensitivity artifacts, which is used for implementing the automatic prompting method for the magnetic resonance magnetic sensitivity artifacts, and comprises the following steps:
an acquisition unit for acquiring a magnetic resonance image and a corresponding magnetic field distribution map;
the model prediction unit is connected with the acquisition unit and used for inputting the magnetic resonance image and the magnetic field distribution map into an artifact prediction model to obtain whether the magnetic resonance image contains magnetic sensitivity artifacts;
and the marking unit is connected with the model prediction unit and used for marking the positions of all the magnetic susceptibility artifacts in response to the magnetic resonance image containing the magnetic susceptibility artifacts and determining the probability of each magnetic susceptibility artifact.
Preferably, the method further comprises the following steps: a model training unit, configured to pre-train the magnetic resonance image and the corresponding magnetic field distribution map as a training data set to obtain the artifact prediction model, where the model training unit includes:
a data amplification module for performing data amplification on the training data set, the data amplification at least comprising: the magnetic resonance image is locally brightened to simulate diseased tissue.
The technical scheme of the invention has the advantages or beneficial effects that:
the invention can predict whether the magnetic resonance image contains the magnetic sensitivity artifact or not through the model, automatically identify the magnetic sensitivity artifact and label and display the magnetic sensitivity artifact in response to the magnetic resonance image containing the magnetic sensitivity artifact, and distinguish the magnetic sensitivity artifact from possible real lesions, thereby effectively reducing the misdiagnosis problem of doctors caused by the artifact.
Drawings
FIG. 1 is a diffusion weighted imaging plot obtained using planar echo acquisition;
FIG. 2 is a magnetic field distribution map corresponding to the tissue region of FIG. 1;
FIG. 3 is a flow chart illustrating a method for automatically prompting MR susceptibility artifacts according to a preferred embodiment of the present invention;
FIG. 4 is a block diagram of an automatic magnetic resonance susceptibility artifact prompt system according to a preferred embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that the embodiments and features of the embodiments may be combined with each other without conflict.
The invention is further described with reference to the following drawings and specific examples, which are not intended to be limiting.
As shown in fig. 1, for Diffusion Weighted Imaging (DWI) acquired by Echo Planar Imaging (EPI), the magnetic sensitivity artifact in the DWI image is particularly prominent, mainly represented by an abnormal highlight signal at the position indicated by an edge arrow, and the highlight signal is easily confused with a lesion.
In a preferred embodiment of the present invention, based on the above problems in the prior art, an automatic prompting method for magnetic resonance magnetic sensitivity artifact is provided, which belongs to the technical field of image processing, and as shown in fig. 3, the method includes:
acquiring a magnetic resonance image and acquiring a corresponding magnetic field distribution map;
inputting the magnetic resonance image and the magnetic field distribution map into an artifact prediction model to obtain whether the magnetic resonance image contains magnetic sensitivity artifacts;
and responding to the magnetic susceptibility artifacts contained in the magnetic resonance image, labeling the positions of all the magnetic susceptibility artifacts, and determining the probability of each magnetic susceptibility artifact.
Specifically, in the embodiment of the present invention, the neural network model (referred to as a deep learning network/model) predicts the position where the magnetic sensitivity artifact may occur according to the magnetic resonance image and the magnetic field distribution map, and the artifact prediction model can distinguish the magnetic sensitivity artifact from a real lesion, assist in determining whether the magnetic resonance image includes the magnetic sensitivity artifact, and mark the position of the magnetic sensitivity artifact when it is determined that the magnetic resonance image includes the magnetic sensitivity artifact, so as to remind a doctor to reduce a misdiagnosis rate.
Further, as shown in fig. 2, which is a magnetic field distribution diagram corresponding to the tissue region of fig. 1, it can be seen that the abnormal highlight signal of fig. 1 shows a normal state in the magnetic field distribution diagram. In embodiments of the invention, the magnetic field pattern is provided as an input to the neural network model. The magnetic field distribution diagram contains physical information of magnetic sensitivity artifacts, and the prediction accuracy of the magnetic sensitivity artifacts can be improved.
As a preferred embodiment, wherein the acquiring of the corresponding magnetic field profile comprises:
acquiring an echo image; and calculating according to the phases of the two echo images to obtain a magnetic field distribution diagram.
In a preferred embodiment, the echo image comprises a dual echo gradient echo image or a two-way phase-coded planar echo image.
Specifically, in a preferred embodiment, a double-echo gradient echo sequence (double-echo gradient echo) can be used for acquisition, and a magnetic field distribution map is obtained by calculation according to the phase difference of two echo images;
in another preferred embodiment, bidirectional phase encoded planar echo (echo) acquisition can be used, and the acquisition can be obtained by using the existing magnetic field pattern calculation method, which is not described herein again.
Furthermore, other existing acquisition modes can be adopted to obtain the magnetic field distribution map. A training data set of the model, i.e. comprising the above-mentioned magnetic resonance image and the above-mentioned magnetic field distribution map, is obtained, for example, by means of a simulation algorithm.
In a preferred embodiment, the artifact prediction model is a deep learning model.
Specifically, in this embodiment, a network is constructed based on a deep learning model, and training is performed according to the network constructed by the training data set, so as to obtain the artifact prediction model according to the embodiment of the present invention. The deep learning model may be implemented using a variety of different architectures. For example, a faster RCNN structure, or a YOLO X structure; alternatively, other mature network architectures may be used, such as efficientDet, YOLO V3/V4/V5, RetinaNet, and the like.
As a preferred embodiment, wherein the inputting the magnetic resonance image and the magnetic field distribution map into an artifact prediction model comprises:
the magnetic resonance image is spliced with the magnetic field distribution map in the channel dimension according to a preset combination mode, and the artifact prediction model is input layer by layer in a two-dimensional form.
As a preferred embodiment, the preset combination includes any one of an absolute value, an absolute value and a phase, and a real part and an imaginary part of the magnetic resonance image.
Specifically, in this embodiment, the magnetic resonance image and the magnetic field distribution map may be input into the artifact prediction model in different combinations, and the specific combination is as follows:
combination mode 1: and acquiring an absolute value of the magnetic resonance image, splicing the absolute value of the magnetic resonance image and the magnetic field distribution map in a channel dimension, and performing artifact prediction in a two-dimensional form layer by layer.
Combination mode 2: and acquiring the absolute value and the phase of the magnetic resonance image, splicing the absolute value and the phase of the magnetic resonance image with the magnetic field distribution map in the channel dimension, and inputting the absolute value and the phase of the magnetic resonance image into the artifact prediction model layer by layer in a two-dimensional form.
Combination mode 3: acquiring a real part and an imaginary part of the magnetic resonance image, splicing the acquired real part and imaginary part of the magnetic resonance image with the magnetic field distribution diagram in a channel dimension, and inputting the spliced magnetic resonance image and the magnetic field distribution diagram into an artifact prediction model layer by layer in a two-dimensional form.
Furthermore, in the three combination methods, the original data for generating the magnetic field distribution map, that is, the dual-echo gradient echo image or the bi-directional phase-encoded plane echo image, may be input into the artifact prediction model, and then spliced with the magnetic resonance image in the channel dimension.
In a preferred embodiment, the artifact prediction model is obtained by pre-training a magnetic resonance image and a corresponding magnetic field distribution map as a training data set, and before training, the artifact prediction model further includes:
and performing data amplification on the training data set.
As a preferred embodiment, wherein the data amplification comprises at least: the magnetic resonance image is locally brightened to simulate diseased tissue.
Specifically, in this embodiment, a local bright signal is added to a part of the magnetic resonance image data of the training data set to simulate the condition of the diseased tissue, so that it can be avoided that the model marks a position where a magnetic sensitivity artifact is likely to be seen as long as the bright signal is seen in the training process, a true diseased region is predicted as the condition of the magnetic sensitivity artifact, and by introducing the simulation of the lesion, the artifact prediction model really learns that the bright signal appears and a region with a large magnetic field distribution change in the magnetic field distribution map is determined as the position where the magnetic sensitivity artifact is likely to be seen, thereby reducing the prediction error rate.
Furthermore, lesion simulation can be realized by changing random brightness change, shape change of lesions and transformation of surface textures; or it may also be assisted by using an anti-generative network (GAN) to approximate the real lesion well by simulating the lesion.
Further, the data amplification method may further include: clipping, translating, changing brightness, adding noise, rotating angle and mirroring to increase the number of samples of the training data set and provide the prediction effect of the model.
Furthermore, in the training process of the model, the output of the model, namely the marking of the position of the magnetic sensitivity artifact, can be manually marked by an experienced doctor; or, the magnetic field distribution map may be automatically labeled according to an algorithm, for example, threshold segmentation is performed on the magnetic field distribution map, and the position where the magnetic field change rate is large is selected as a possible position of the magnetic sensitivity artifact, and then labeled.
In the above preferred embodiment, the artifact prediction model is trained as follows:
preparing a training data set: the input data required by the model mainly comprises two parts, wherein the first part is data needing prediction, namely magnetic resonance image data possibly having magnetic sensitivity artifacts; the second is the current magnetic field map (field map). The magnetic resonance image data can be obtained from a magnetic resonance scanner and can be acquired from healthy volunteers through a plane echo sequence or a gradient echo sequence according to different target application scenes, and the specific acquisition sequence can be determined according to actual use scenes and possibly suspected lesion types; the magnetic field profile may be obtained by a dual echo gradient echo sequence or a two-way phase encoded echo planar sequence. The weights of the neural network are optimized by collecting a certain amount of training data sets so as to improve the prediction effect of the model.
Training a network: and (3) randomly initializing the weight of the deep learning model, and training the deep learning model through a prepared training data set to obtain a trained artifact prediction model.
Artifact prediction: in practical use, required data to be predicted are collected and input into a trained artifact prediction model according to a combination mode defined by the trained artifact prediction model, namely, the position and the probability of a predicted magnetic sensitivity artifact can be output, and the position of the magnetic sensitivity artifact is marked in a magnetic resonance image by a pattern with a preset shape, wherein the preset shape comprises a rectangular frame, an arrow, a circle and a square frame.
The present invention further provides an automatic prompting system for magnetic resonance magnetic susceptibility artifact, which is used for implementing the above automatic prompting method for magnetic resonance magnetic susceptibility artifact, as shown in fig. 4, and includes:
an acquisition unit 1 for acquiring a magnetic resonance image and a corresponding magnetic field distribution map;
the model prediction unit 2 is connected with the acquisition unit 1 and is used for inputting the magnetic resonance image and the magnetic field distribution map into an artifact prediction model to obtain whether the magnetic resonance image contains magnetic sensitivity artifacts;
and the labeling unit 3 is connected with the model prediction unit 2 and used for labeling the positions of all the magnetic susceptibility artifacts in response to the magnetic resonance image containing the magnetic susceptibility artifacts and determining the probability of each magnetic susceptibility artifact.
As a preferred embodiment, the method further comprises the following steps: a model training unit 4, configured to pre-train the magnetic resonance image and the corresponding magnetic field distribution map as a training data set to obtain an artifact prediction model, where the model training unit 4 includes:
a data amplification module 41, configured to perform data amplification on the training data set, where the data amplification at least includes: the magnetic resonance image is locally brightened to simulate diseased tissue.
The technical scheme of the invention has the following advantages or beneficial effects: the invention can predict whether the magnetic resonance image contains the magnetic sensitivity artifact or not through the model, automatically identify the magnetic sensitivity artifact and label and display the magnetic sensitivity artifact in response to the magnetic resonance image containing the magnetic sensitivity artifact, and distinguish the magnetic sensitivity artifact from possible real lesions, thereby effectively reducing the misdiagnosis problem of doctors caused by the artifact.
While the invention has been described with reference to a preferred embodiment, it will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the spirit and scope of the invention.
Claims (10)
1. An automatic prompting method for magnetic resonance magnetic sensitivity artifacts is characterized by comprising the following steps:
acquiring a magnetic resonance image and acquiring a corresponding magnetic field distribution map;
inputting the magnetic resonance image and the magnetic field distribution map into an artifact prediction model to obtain whether the magnetic resonance image contains magnetic sensitivity artifacts;
and responding to the magnetic resonance image containing magnetic susceptibility artifacts, labeling the positions of all the magnetic susceptibility artifacts, and determining the probability of each magnetic susceptibility artifact.
2. The method of claim 1, wherein obtaining a corresponding magnetic field distribution map comprises:
acquiring an echo image; and calculating according to the phases of the two echo images to obtain the magnetic field distribution diagram.
3. The method of claim 2, wherein the echo image comprises a dual echo gradient echo image or a bi-directional phase encoded plane echo image.
4. The method of claim 1, wherein the artifact prediction model is a deep learning model.
5. The method of claim 1, wherein inputting the magnetic resonance image and the magnetic field distribution map into an artifact prediction model comprises:
and splicing the magnetic resonance image and the magnetic field distribution map in a channel dimension according to a preset combination mode, and inputting the artifact prediction model layer by layer in a two-dimensional form.
6. The method of claim 5, wherein the predetermined combination comprises any one of absolute value, absolute value and phase, real part and imaginary part of the magnetic resonance image.
7. The method of claim 1, wherein the artifact prediction model is pre-trained using the magnetic resonance image and the corresponding magnetic field distribution map as a training data set, and further comprises, before training:
performing data amplification on the training data set.
8. The method of claim 1, wherein the data augmentation comprises at least: the magnetic resonance image is locally brightened to simulate diseased tissue.
9. An automatic prompting system for magnetic resonance magnetic susceptibility artifact, for implementing the automatic prompting method for magnetic resonance magnetic susceptibility artifact according to any one of claims 1-8, comprising:
an acquisition unit for acquiring a magnetic resonance image and a corresponding magnetic field distribution map;
the model prediction unit is connected with the acquisition unit and used for inputting the magnetic resonance image and the magnetic field distribution map into an artifact prediction model to obtain whether the magnetic resonance image contains magnetic sensitivity artifacts;
and the marking unit is connected with the model prediction unit and used for marking the positions of all the magnetic susceptibility artifacts in response to the magnetic resonance image containing the magnetic susceptibility artifacts and determining the probability of each magnetic susceptibility artifact.
10. The system of claim 9, further comprising: a model training unit, configured to pre-train the magnetic resonance image and the corresponding magnetic field distribution map as a training data set to obtain the artifact prediction model, where the model training unit includes:
a data amplification module for performing data amplification on the training data set, the data amplification at least comprising: the magnetic resonance image is locally brightened to simulate diseased tissue.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210735986.1A CN115100310A (en) | 2022-06-27 | 2022-06-27 | Automatic prompting method and system for magnetic resonance magnetic sensitivity artifact |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210735986.1A CN115100310A (en) | 2022-06-27 | 2022-06-27 | Automatic prompting method and system for magnetic resonance magnetic sensitivity artifact |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115100310A true CN115100310A (en) | 2022-09-23 |
Family
ID=83293917
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210735986.1A Pending CN115100310A (en) | 2022-06-27 | 2022-06-27 | Automatic prompting method and system for magnetic resonance magnetic sensitivity artifact |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115100310A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115359144A (en) * | 2022-10-19 | 2022-11-18 | 之江实验室 | Magnetic resonance plane echo imaging artifact simulation method and system |
| CN117409016A (en) * | 2023-12-15 | 2024-01-16 | 华中科技大学同济医学院附属同济医院 | Automatic segmentation method for magnetic resonance image |
-
2022
- 2022-06-27 CN CN202210735986.1A patent/CN115100310A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115359144A (en) * | 2022-10-19 | 2022-11-18 | 之江实验室 | Magnetic resonance plane echo imaging artifact simulation method and system |
| CN115359144B (en) * | 2022-10-19 | 2023-03-03 | 之江实验室 | Magnetic resonance plane echo imaging artifact simulation method and system |
| CN117409016A (en) * | 2023-12-15 | 2024-01-16 | 华中科技大学同济医学院附属同济医院 | Automatic segmentation method for magnetic resonance image |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109754387B (en) | Intelligent detection and positioning method for whole-body bone imaging radioactive concentration focus | |
| US9858665B2 (en) | Medical imaging device rendering predictive prostate cancer visualizations using quantitative multiparametric MRI models | |
| CN111428709B (en) | Image processing method, device, computer equipment and storage medium | |
| US20190114773A1 (en) | Systems and methods for cross-modality image segmentation | |
| Zhang et al. | Automated deep learning method for whole‐breast segmentation in diffusion‐weighted breast MRI | |
| Bologna et al. | Virtual phantom analyses for preprocessing evaluation and detection of a robust feature set for MRI‐radiomics of the brain | |
| CN115100310A (en) | Automatic prompting method and system for magnetic resonance magnetic sensitivity artifact | |
| CN106999091B (en) | Method and system for improved classification of constituent materials | |
| CN108629785B (en) | Three-dimensional magnetic resonance pancreas image segmentation method based on self-learning | |
| CN102999917B (en) | Cervical cancer image automatic segmentation method based on T2-MRI and DW-MRI | |
| CN107644420A (en) | Blood-vessel image dividing method, MRI system based on central line pick-up | |
| CN112950639B (en) | SA-Net-based MRI medical image segmentation method | |
| Fernandes et al. | Prostate fiducial marker detection with the use of multi-parametric magnetic resonance imaging | |
| Delbany et al. | One‐millimeter isotropic breast diffusion‐weighted imaging: Evaluation of a superresolution strategy in terms of signal‐to‐noise ratio, sharpness and apparent diffusion coefficient | |
| CN116188479B (en) | Hip joint image segmentation method and system based on deep learning | |
| CN112802025A (en) | Liver tumor segmentation method and device under CT image | |
| Bueno et al. | Three-dimensional segmentation of anatomical structures in MR images on large data bases | |
| Dwyer et al. | DeepGRAI (deep gray rating via artificial intelligence): fast, feasible, and clinically relevant thalamic atrophy measurement on clinical quality T2-FLAIR MRI in multiple sclerosis | |
| CN111784652B (en) | MRI (magnetic resonance imaging) segmentation method based on reinforcement learning multi-scale neural network | |
| Chang | Improving the Otsu method for MRA image vessel extraction via resampling and ensemble learning | |
| CN116309551B (en) | Method, device, electronic equipment and readable medium for determining focus sampling area | |
| Li et al. | Breast MRI segmentation and Ki-67 high-and low-expression prediction algorithm based on deep learning | |
| WO2023202662A1 (en) | Cest magnetic resonance analysis method based on permuted random forest, and apparatus | |
| Mathai et al. | Universal detection and segmentation of lymph nodes in multi-parametric MRI | |
| Clèrigues et al. | Minimizing the effect of white matter lesions on deep learning based tissue segmentation for brain volumetry |
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 |