CN112330674B - Self-adaptive variable-scale convolution kernel method based on brain MRI three-dimensional image confidence coefficient - Google Patents
Self-adaptive variable-scale convolution kernel method based on brain MRI three-dimensional image confidence coefficient Download PDFInfo
- Publication number
- CN112330674B CN112330674B CN202011441419.2A CN202011441419A CN112330674B CN 112330674 B CN112330674 B CN 112330674B CN 202011441419 A CN202011441419 A CN 202011441419A CN 112330674 B CN112330674 B CN 112330674B
- Authority
- CN
- China
- Prior art keywords
- convolution kernel
- calculating
- coding direction
- confidence
- confidence coefficient
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60L—PROPULSION OF ELECTRICALLY-PROPELLED VEHICLES; SUPPLYING ELECTRIC POWER FOR AUXILIARY EQUIPMENT OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRODYNAMIC BRAKE SYSTEMS FOR VEHICLES IN GENERAL; MAGNETIC SUSPENSION OR LEVITATION FOR VEHICLES; MONITORING OPERATING VARIABLES OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRIC SAFETY DEVICES FOR ELECTRICALLY-PROPELLED VEHICLES
- B60L3/00—Electric devices on electrically-propelled vehicles for safety purposes; Monitoring operating variables, e.g. speed, deceleration or energy consumption
- B60L3/0023—Detecting, eliminating, remedying or compensating for drive train abnormalities, e.g. failures within the drive train
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60L—PROPULSION OF ELECTRICALLY-PROPELLED VEHICLES; SUPPLYING ELECTRIC POWER FOR AUXILIARY EQUIPMENT OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRODYNAMIC BRAKE SYSTEMS FOR VEHICLES IN GENERAL; MAGNETIC SUSPENSION OR LEVITATION FOR VEHICLES; MONITORING OPERATING VARIABLES OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRIC SAFETY DEVICES FOR ELECTRICALLY-PROPELLED VEHICLES
- B60L3/00—Electric devices on electrically-propelled vehicles for safety purposes; Monitoring operating variables, e.g. speed, deceleration or energy consumption
- B60L3/12—Recording operating variables ; Monitoring of operating variables
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/082—Learning methods modifying the architecture, e.g. adding, deleting or silencing nodes or connections
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T9/00—Image coding
-
- 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]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30016—Brain
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/30—Assessment of water resources
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Mechanical Engineering (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Sustainable Development (AREA)
- Sustainable Energy (AREA)
- Power Engineering (AREA)
- Transportation (AREA)
- Multimedia (AREA)
- Medical Informatics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Magnetic Resonance Imaging Apparatus (AREA)
Abstract
The invention discloses a brain MRI three-dimensional image confidence coefficient-based self-adaptive variable scale convolution kernel method, which belongs to the technical field of information and automatic control, and aims to automatically change the size of convolution kernels in real time so as to better extract features, wherein the confidence coefficient of MRI images in frequency coding, phase coding and layer selection coding directions corresponds to the fact that if the confidence coefficient of a tumor area is high, more and more convolution kernels are provided for extracting features, otherwise, small and fewer convolution kernels are used for extracting features, so that calculation resources are used at key positions, and the problem that gray scale ranges of images obtained by different patients under different instruments are different is solved. The MRI tumor image is better segmented.
Description
Technical Field
The invention belongs to the technical field of information and automatic control, and particularly relates to a self-adaptive variable-scale convolution kernel method based on brain MRI three-dimensional image confidence.
Background
The mortality rate of brain tumors is high, and once diagnosed, patients typically remain for two or three years. Magnetic Resonance Imaging (MRI) techniques have been widely used for imaging decisions of various systems of the human body, whereby doctors diagnose the disease through magnetic resonance images so as to discover the disease and treat it as early as possible.
The rapid, automatic and accurate detection of brain MRI tumor images is of vital importance for tumor diagnosis. While many deep learning networks such as Full Convolution Networks (FCNs) are now able to detect tumor regions in MRI images of the brain, since the size and number of convolution layers in such networks are fixed, extracting features by such invariant convolution kernels necessarily loses some detailed information after passing through multiple convolution layers or extracts much useless information, wasting computational resources. The method can not achieve the rapidness and accuracy in the automatic brain MRI tumor segmentation, and can even influence doctors to make wrong judgment.
Disclosure of Invention
The invention aims to solve the technical problem of a self-adaptive variable-scale convolution kernel method based on brain MRI three-dimensional image confidence. Because of different gray scale ranges of images obtained by different patients and different instruments, the confidence coefficient of MRI image targets is also different, and the method provided by the invention can automatically adjust the scale of convolution kernels in a depth network according to the confidence coefficient of MRI image tumors.
The invention adopts the following technical scheme for solving the technical problems:
an adaptive variable-scale convolution kernel method based on brain MRI three-dimensional image confidence comprises two parts of calculation of target region confidence in an MRI image and calculation of an adaptive variable-scale convolution kernel;
the confidence of the target area in the MRI image is calculated as follows:
step 1: the signal-to-noise ratio SNR is calculated as follows:
step 1.1, measuring a measurement interest region MROI of a central signal region of the image from x, y and z directions on the image uniformity layer respectively to obtain signal intensity S x ,S y ,S z ;
Step 1.2, measuring frequency codes and phase codes respectively from four corner background areas around the die body, wherein the area of the layer selection coding direction is 100mm 2 Standard deviation of ROI signal intensity for a region of interestFurther calculating the signal-to-noise ratio SNR;
step 2, calculating the uniformity of the main magnetic field;
step 2.1, calculating the linear gradient fields G for frequency, phase and slice selection, respectively x ,G y And G z The method comprises the steps of carrying out a first treatment on the surface of the Concrete embodimentsThe following are provided:
wherein gamma is gyromagnetic ratio, FOV x ,FOV y And FOV (field of view) z Effective Field of view (BW) along the frequency coding direction, the phase coding direction and the layer selection coding direction, respectively x ,BW y And BW z The receiving bandwidths of the coding directions are selected along the frequency coding direction, the phase coding direction and the layer plane respectively;
step 2.2, calculating the magnetic field DeltaB which is generated in the frequency coding direction, the phase coding direction and the layer selection coding direction and is caused by local non-uniform 0 Image distortion amount caused by (x, y, z):
x’=x+ΔB 0x (x,y,z)/G x
y’=y+ΔB 0y (x,y,z)/G y
z’=z+ΔB 0z (x,y,z)/G z
step 2.3, for the uniformity of the spherical model magnetic field, calculating the main magnetic field uniformity along the frequency coding, the phase coding and the layer selection coding directions respectively by using the following steps:
wherein BW is x1 ,BW y1 ,BW z1 Respectively smaller reception bandwidths; BW (BW) x2 ,BW y2 ,BW z2 Respectively, larger receiving bandwidth, x' 1 -x’ 2 ,y’ 1 -y’ 2 ,z’ 1 -z’ 2 The displacement difference value of the deformation of the image obtained by 2 times of scanning of the higher bandwidth in the respective coding direction is respectively obtained in the smaller bandwidth; b (B) 0 (T) is the main magnetic field strength;
step 2.4, calculating confidence coefficients on the x, y and z axes respectively according to the steps, namely calculating the confidence coefficients of the K space frequency coding direction, the phase coding direction and the layer selection coding direction of the corresponding MRI; the confidence in three directions is defined as:
P x =βSNR x +(1-β)ΔB 0x (ppm)
P y =βSNR y +(1-β)ΔB 0y (ppm)
P z =βSNR z +(1-β)ΔB 0z (ppm)
wherein β is a weight parameter set by man, and β=0.5;
the calculation of the adaptive variable-scale convolution kernel is specifically as follows:
step 3, calculating the specific multi-scale convolution kernel size according to the calculated confidence coefficient of the target object;
step 3.1, finding out the relation between the confidence coefficient of the target object and the convolution kernel scale; wherein the size of the three-dimensional convolution kernel is denoted as w, h, d; wherein, the initial values of w, h and d are all 3, and are marked as w 0 ,h 0 ,d 0 The method comprises the steps of carrying out a first treatment on the surface of the Since the confidence value is a fraction between 0 and 1, it is not suitable to be used directly as a convolution kernel scale; the size of the three-dimensional convolution kernel is determined by the following equation:
performing upward rounding processing on the decimal parts appearing in the three formulas;
based on the two groups of formulas, the three-dimensional convolution kernel has self-adaptive capability for various different targets, namely, the three-dimensional convolution kernel can automatically adjust the scale of the three-dimensional convolution kernel according to the confidence coefficient on the space of the different targets and the probability of the occurrence of the targets.
Compared with the prior art, the technical scheme provided by the invention has the following technical effects:
the self-adaptive variable-scale convolution kernel method based on the confidence coefficient of the brain MRI image can automatically change the scale of the convolution kernel in real time according to the confidence coefficient of a tumor area in the MRI image so as to better extract the characteristics, if the confidence coefficient of the tumor area is high, a large scale convolution kernel is provided to extract the characteristics, otherwise, a small convolution kernel is used to extract the characteristics, so that the calculation resources are used in key positions, the problem that the gray scale range of the MRI image obtained by different patients under different instruments is different is solved, the size of the convolution kernel is changed in real time according to the confidence coefficient of a three-dimensional image pixel, and the MRI tumor image can be segmented better.
Drawings
FIG. 1 is a schematic illustration of an MRI three-dimensional image homogeneity layer in accordance with an embodiment of the present invention;
fig. 2 is a schematic diagram of an adaptive variable scale convolution kernel in accordance with an embodiment of the present invention.
Detailed Description
The technical scheme of the invention is further described in detail below with reference to the accompanying drawings:
the following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The invention provides a self-adaptive variable-scale convolution kernel method based on brain MRI three-dimensional image confidence coefficient based on the existing artificial intelligence technology.
Because of different quality of MRI images obtained under different instruments of different patients, confidence of targets in the images can be different, and three-dimensional convolution kernels with different scales are needed for better extracting critical features.
The technical solution for realizing the purpose of the invention is as follows: an adaptive variable-scale convolution kernel method based on the confidence coefficient of a brain MRI three-dimensional image is particularly used for changing the size of a convolution kernel in real time according to the confidence coefficient value of a region of interest in the MRI three-dimensional image.
Module one: the method for calculating the confidence of the MRI image is as follows:
step 1: the signal-to-noise ratio (SNR) is calculated.
Step 1.1: as shown in fig. 1, the measurement interest area (MROI) of the central signal area of the image is measured (at least 75% of a centered regular geometric area) from the x, y, z directions on the image uniformity layer, respectively, to obtain the signal intensity: s is S x ,S y ,S z :
Step 1.1.1: measuring frequency code, phase code and layer selection code direction area of 100mm from four corner background areas around the die body 2 Standard deviation of region of interest (ROI) signal intensityThe signal-to-noise ratio (SNR) is calculated from the following equation:
step 1.2: and calculating the uniformity of the main magnetic field.
Step 1.2.1: calculating the frequency, phase and slice-selective linear gradient field G, respectively x ,G y And G z ;
Wherein gamma is gyromagnetic ratio, FOV x ,FOV y And FOV (field of view) z Effective Field of view (BW) along the frequency coding direction, the phase coding direction and the layer selection coding direction, respectively x ,BW y And BW z The reception bandwidths of the coding directions are selected along the frequency coding direction, the phase coding direction and the layer plane, respectively.
Step 1.2.2: calculating the magnetic field delta B caused by local non-uniformity in the frequency coding direction, the phase coding direction and the layer selection coding direction 0 Image distortion caused by (x, y, z)
x’=x+ΔB 0x (x,y,z)/G x
y’=y+ΔB 0y (x,y,z)/G y
z’=z+ΔB 0z (x,y,z)/G z
Step 1.2.3: the uniformity of the spherical model magnetic field is calculated along the frequency coding and the phase coding respectively by using the following steps:
wherein BW is x1 ,BW y1 ,BW z1 Which are respectively smaller reception bandwidths. BW (BW) x2 ,BW y2 ,BW z2 Respectively, a larger receiving bandwidth, (x' 1 -x’ 2 ),(y’ 1 -y’ 2 ),(z’ 1 -z’ 2 ) The displacement difference of the deformation of the image in the respective coding direction obtained by 2 scans of the higher bandwidth is respectively smaller bandwidth. B (B) 0 And (T) is the main magnetic field strength.
Step 1.2.4: and respectively calculating confidence degrees on x, y and z axes according to the steps, namely calculating the confidence degrees of the K space frequency coding direction, the phase coding direction and the layer selection coding direction of the corresponding MRI. This patent defines the confidence in three directions as:
P x =βSNR x +(1-β)ΔB 0x (ppm)
P y =βSNR y +(1-β)ΔB 0y (ppm)
P z =βSNR z +(1-β)ΔB 0z (ppm)
where β is a weight parameter set by man, β=0.5 is set in this patent.
And a second module: the calculation of the adaptive variable-scale convolution kernel is specifically as follows:
step 1: according to the first calculation of the moduleThe confidence of the calculated target object is used for solving the specific multi-scale convolution kernel size. First, the relation between the confidence of the target object and the convolution kernel scale is found. The size of the three-dimensional convolution kernel is expressed as (w, h, d), the initial values of w, h and d are all 3, and the initial values are marked as w 0 ,h 0 ,d 0 . Since the confidence value is a fraction between 0 and 1, it is not suitable to be directly used as a convolution kernel scale. The size of the three-dimensional convolution kernel is determined by the following formula:
the fractional part appearing in the above three formulas is rounded up.
Based on the two groups of formulas, the three-dimensional convolution kernel has self-adaptive capability for various different targets, namely, the three-dimensional convolution kernel can automatically adjust the scale of the three-dimensional convolution kernel according to the confidence coefficient on the space of the different targets and the probability of the occurrence of the targets.
It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the prior art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.
The above embodiments are only for illustrating the technical idea of the present invention, and the protection scope of the present invention is not limited thereto, and any modification made on the basis of the technical scheme according to the technical idea of the present invention falls within the protection scope of the present invention. The embodiments of the present invention have been described in detail, but the present invention is not limited to the above embodiments, and various changes can be made within the knowledge of those skilled in the art without departing from the spirit of the present invention.
Claims (1)
1. An adaptive variable-scale convolution kernel method based on brain MRI three-dimensional image confidence is characterized in that: the method comprises the steps of calculating the confidence coefficient of a target area in an MRI image and calculating an adaptive variable-scale convolution kernel;
the confidence of the target area in the MRI image is calculated as follows:
step 1: the signal-to-noise ratio SNR is calculated as follows:
step 1.1, measuring a measurement interest region MROI of a central signal region of the image from x, y and z directions on the image uniformity layer respectively to obtain signal intensity S x ,S y ,S z ;
Step 1.2, measuring frequency codes and phase codes respectively from four corner background areas around the die body, wherein the area of the layer selection coding direction is 100mm 2 Standard deviation of ROI signal intensity for a region of interestFurther calculating the signal-to-noise ratio SNR;
step 2, calculating the uniformity of the main magnetic field;
step 2.1, calculating the linear gradient fields G for frequency, phase and slice selection, respectively x ,G y And G z The method comprises the steps of carrying out a first treatment on the surface of the The method comprises the following steps:
wherein gamma is gyromagnetic ratio, FOV x ,FOV y And FOV (field of view) z The effective Field of view, BW of the coding direction is selected along the frequency coding direction, the phase coding direction and the layer respectively x ,BW y And BW z The receiving bandwidths of the coding directions are selected along the frequency coding direction, the phase coding direction and the layer plane respectively;
step 2.2, calculating the magnetic field DeltaB which is generated in the frequency coding direction, the phase coding direction and the layer selection coding direction and is caused by local non-uniform 0 Image distortion amount caused by (x, y, z):
x′=x+ΔB 0x (x,y,z)/G x
y′=y+ΔB 0y (x,y,z)/G y
z′=z+ΔB 0z (x,y,z)/G z
step 2.3, for the uniformity of the spherical model magnetic field, calculating the main magnetic field uniformity along the frequency coding, the phase coding and the layer selection coding directions respectively by using the following steps:
wherein BW is x1 ,BW y1 ,BW z1 Respectively smaller reception bandwidths; BW (BW) x2 ,BW y2 ,BW z2 Respectively, larger receiving bandwidth, x' 1 -x′ 2 ,y′ 1 -y′ 2 ,z′ 1 -z′ 2 The displacement difference value of the deformation of the image obtained by 2 times of scanning of the higher bandwidth in the respective coding direction is respectively obtained in the smaller bandwidth; b (B) 0 (T) is the main magnetic field strength;
step 2.4, calculating confidence coefficients on the x, y and z axes respectively according to the steps, namely calculating the confidence coefficients of the K space frequency coding direction, the phase coding direction and the layer selection coding direction of the corresponding MRI; the confidence in three directions is defined as:
P x =βSNR x +(1-β)ΔB 0x (ppm)
P y =βSNR y +(1-β)ΔB 0y (ppm)
P z =βSNR z +(1-β)ΔB 0z (ppm)
wherein β is a weight parameter set by man, and β=0.5;
the calculation of the adaptive variable-scale convolution kernel is specifically as follows:
step 3, calculating the specific multi-scale convolution kernel size according to the calculated confidence coefficient of the target object;
step 3.1, finding out the relation between the confidence coefficient of the target object and the convolution kernel scale; wherein the size of the three-dimensional convolution kernel is denoted as w, h, d; wherein, the initial values of w, h and d are all 3, and are marked as w 0 ,h 0 ,d 0 The method comprises the steps of carrying out a first treatment on the surface of the The size of the three-dimensional convolution kernel is then determined by the following formula:
performing upward rounding processing on the decimal parts appearing in the three formulas;
based on the two groups of formulas, the three-dimensional convolution kernel has self-adaptive capability for various different targets, namely, the three-dimensional convolution kernel can automatically adjust the scale of the three-dimensional convolution kernel according to the confidence coefficient on the space of the different targets and the probability of the occurrence of the targets.
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010376917 | 2020-05-07 | ||
CN2020103769177 | 2020-05-07 |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112330674A CN112330674A (en) | 2021-02-05 |
CN112330674B true CN112330674B (en) | 2023-06-30 |
Family
ID=74301865
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011441419.2A Active CN112330674B (en) | 2020-05-07 | 2020-12-08 | Self-adaptive variable-scale convolution kernel method based on brain MRI three-dimensional image confidence coefficient |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112330674B (en) |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107064838A (en) * | 2017-04-25 | 2017-08-18 | 北京青檬艾柯科技有限公司 | It is a kind of to form the magnet system configurations and measuring method for becoming gradient magnetostatic field |
CN107240102A (en) * | 2017-04-20 | 2017-10-10 | 合肥工业大学 | Malignant tumour area of computer aided method of early diagnosis based on deep learning algorithm |
CN107403201A (en) * | 2017-08-11 | 2017-11-28 | 强深智能医疗科技(昆山)有限公司 | Tumour radiotherapy target area and jeopardize that organ is intelligent, automation delineation method |
CN109086770A (en) * | 2018-07-25 | 2018-12-25 | 成都快眼科技有限公司 | A kind of image, semantic dividing method and model based on accurate scale prediction |
CN109872364A (en) * | 2019-01-28 | 2019-06-11 | 腾讯科技(深圳)有限公司 | Image-region localization method, device, storage medium and medical image processing equipment |
CN109886929A (en) * | 2019-01-24 | 2019-06-14 | 江苏大学 | A kind of MRI tumour voxel detection method based on convolutional neural networks |
CN109978838A (en) * | 2019-03-08 | 2019-07-05 | 腾讯科技(深圳)有限公司 | Image-region localization method, device and Medical Image Processing equipment |
CN110415234A (en) * | 2019-07-29 | 2019-11-05 | 北京航空航天大学 | Brain tumor dividing method based on multi-parameter magnetic resonance imaging |
CN111027547A (en) * | 2019-12-06 | 2020-04-17 | 南京大学 | Automatic detection method for multi-scale polymorphic target in two-dimensional image |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE102016213042A1 (en) * | 2016-07-18 | 2018-01-18 | Siemens Healthcare Gmbh | Method for recording calibration data for GRAPPA algorithms |
JP2020510463A (en) * | 2017-01-27 | 2020-04-09 | アーテリーズ インコーポレイテッド | Automated segmentation using full-layer convolutional networks |
-
2020
- 2020-12-08 CN CN202011441419.2A patent/CN112330674B/en active Active
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107240102A (en) * | 2017-04-20 | 2017-10-10 | 合肥工业大学 | Malignant tumour area of computer aided method of early diagnosis based on deep learning algorithm |
CN107064838A (en) * | 2017-04-25 | 2017-08-18 | 北京青檬艾柯科技有限公司 | It is a kind of to form the magnet system configurations and measuring method for becoming gradient magnetostatic field |
CN107403201A (en) * | 2017-08-11 | 2017-11-28 | 强深智能医疗科技(昆山)有限公司 | Tumour radiotherapy target area and jeopardize that organ is intelligent, automation delineation method |
CN109086770A (en) * | 2018-07-25 | 2018-12-25 | 成都快眼科技有限公司 | A kind of image, semantic dividing method and model based on accurate scale prediction |
CN109886929A (en) * | 2019-01-24 | 2019-06-14 | 江苏大学 | A kind of MRI tumour voxel detection method based on convolutional neural networks |
CN109872364A (en) * | 2019-01-28 | 2019-06-11 | 腾讯科技(深圳)有限公司 | Image-region localization method, device, storage medium and medical image processing equipment |
CN109978838A (en) * | 2019-03-08 | 2019-07-05 | 腾讯科技(深圳)有限公司 | Image-region localization method, device and Medical Image Processing equipment |
CN110415234A (en) * | 2019-07-29 | 2019-11-05 | 北京航空航天大学 | Brain tumor dividing method based on multi-parameter magnetic resonance imaging |
CN111027547A (en) * | 2019-12-06 | 2020-04-17 | 南京大学 | Automatic detection method for multi-scale polymorphic target in two-dimensional image |
Non-Patent Citations (3)
Title |
---|
Efficient multi-scale 3D CNN with fully connected CRF for accurate brain lesion segmentation;Konstantinos Kamnitsas等;《Medical Image Analysis》;第36卷;第61-78页 * |
利用DCE-MRI结合改进卷积神经网络的MR图像自动分割与分类方法;杨珍等;《重庆理工大学学报(自然科学)》;第34卷(第2期);第147-157页 * |
多模态视网膜图像血管分割及配准研究;李苹;《中国优秀硕士学位论文全文数据库 信息科技辑》(第1期);第I138-4604页 * |
Also Published As
Publication number | Publication date |
---|---|
CN112330674A (en) | 2021-02-05 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108776969B (en) | Breast ultrasound image tumor segmentation method based on full convolution network | |
US5529069A (en) | Magnetic resonance imaging apparatus | |
CN111047629B (en) | Multi-modal image registration method and device, electronic equipment and storage medium | |
Chen et al. | MRI brain tissue classification using unsupervised optimized extenics-based methods | |
CN116188496B (en) | Remote sensing image self-adaptive segmentation method based on land utilization type | |
CN104380340A (en) | Scoring and adjusting pixels based on neighborhood relationships for revealing data in images | |
CN103632345A (en) | MRI image inhomogeneity correction method based on regularization | |
Haq et al. | DCNNBT: A novel deep convolution neural network-based brain tumor classification model | |
CN113628234B (en) | Method for detecting salient polarization SAR image change based on comprehensive neighborhood information | |
Wang et al. | Detection of spectral signatures in multispectral MR images for classification | |
CN114140341A (en) | Magnetic resonance image non-uniform field correction method based on deep learning | |
Li et al. | Entropy-based global and local weight adaptive image segmentation models | |
CN110458842B (en) | Brain tumor segmentation method based on two-channel three-dimensional dense connection network | |
CN103871035B (en) | Image denoising method and device | |
CN112330674B (en) | Self-adaptive variable-scale convolution kernel method based on brain MRI three-dimensional image confidence coefficient | |
CN117764864A (en) | nuclear magnetic resonance tumor visual detection method based on image denoising | |
CN111429404B (en) | Imaging system and method for detecting cardiovascular and cerebrovascular vessels | |
CN116503426B (en) | Ultrasonic image segmentation method based on image processing | |
CN107690587B (en) | MRI with variable density sampling | |
CN102737388B (en) | Image processing apparatus and image processing method | |
Anand Kumar et al. | 3D deep learning for automatic brain MR tumor segmentation with T-spline intensity inhomogeneity correction | |
JP6843706B2 (en) | Magnetic resonance imaging device and diffusion-weighted image correction method | |
Arora et al. | Noise adaptive FCM algorithm for segmentation of MRI brain images using local and non-local spatial information | |
Aetesam et al. | Perceptually motivated generative model for magnetic resonance image denoising | |
Thurnhofer-Hemsi et al. | Super-resolution of 3D magnetic resonance images by random shifting and convolutional neural networks |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |