CN117994322A - Automatic fat quantification method, system and equipment based on vertebral body segmentation - Google Patents
Automatic fat quantification method, system and equipment based on vertebral body segmentation Download PDFInfo
- Publication number
- CN117994322A CN117994322A CN202410191127.XA CN202410191127A CN117994322A CN 117994322 A CN117994322 A CN 117994322A CN 202410191127 A CN202410191127 A CN 202410191127A CN 117994322 A CN117994322 A CN 117994322A
- Authority
- CN
- China
- Prior art keywords
- fat
- volume
- area
- vertebral body
- segmentation
- 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
- 230000011218 segmentation Effects 0.000 title claims abstract description 80
- 238000000034 method Methods 0.000 title claims abstract description 63
- 238000011002 quantification Methods 0.000 title claims description 26
- 210000000689 upper leg Anatomy 0.000 claims description 67
- 210000000038 chest Anatomy 0.000 claims description 47
- 210000001015 abdomen Anatomy 0.000 claims description 42
- 210000004003 subcutaneous fat Anatomy 0.000 claims description 42
- 238000012935 Averaging Methods 0.000 claims description 35
- 210000001596 intra-abdominal fat Anatomy 0.000 claims description 34
- 210000000577 adipose tissue Anatomy 0.000 claims description 32
- 210000001185 bone marrow Anatomy 0.000 claims description 28
- 238000004364 calculation method Methods 0.000 claims description 25
- 230000003187 abdominal effect Effects 0.000 claims description 24
- 210000000115 thoracic cavity Anatomy 0.000 claims description 24
- 210000001370 mediastinum Anatomy 0.000 claims description 21
- 210000004197 pelvis Anatomy 0.000 claims description 14
- 210000000579 abdominal fat Anatomy 0.000 claims description 10
- 210000001835 viscera Anatomy 0.000 claims description 9
- 210000003141 lower extremity Anatomy 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 5
- 210000002758 humerus Anatomy 0.000 claims description 5
- 238000004422 calculation algorithm Methods 0.000 claims description 4
- 238000002372 labelling Methods 0.000 claims description 4
- 230000009278 visceral effect Effects 0.000 claims description 4
- 239000012071 phase Substances 0.000 description 22
- RGCLLPNLLBQHPF-HJWRWDBZSA-N phosphamidon Chemical compound CCN(CC)C(=O)C(\Cl)=C(/C)OP(=O)(OC)OC RGCLLPNLLBQHPF-HJWRWDBZSA-N 0.000 description 12
- 238000010586 diagram Methods 0.000 description 7
- 238000003709 image segmentation Methods 0.000 description 6
- 238000012549 training Methods 0.000 description 6
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 6
- 201000010099 disease Diseases 0.000 description 4
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 230000004807 localization Effects 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- 238000005070 sampling Methods 0.000 description 4
- 210000001519 tissue Anatomy 0.000 description 4
- 238000012795 verification Methods 0.000 description 4
- 238000013527 convolutional neural network Methods 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 125000004435 hydrogen atom Chemical group [H]* 0.000 description 3
- 210000000056 organ Anatomy 0.000 description 3
- 238000011176 pooling Methods 0.000 description 3
- 210000004490 abdominal subcutaneous fat Anatomy 0.000 description 2
- 230000004913 activation Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 210000000988 bone and bone Anatomy 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 238000013136 deep learning model Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 210000003734 kidney Anatomy 0.000 description 2
- 210000004185 liver Anatomy 0.000 description 2
- 238000002595 magnetic resonance imaging Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 210000000496 pancreas Anatomy 0.000 description 2
- 238000000926 separation method Methods 0.000 description 2
- 210000000952 spleen Anatomy 0.000 description 2
- 238000007920 subcutaneous administration Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 1
- 208000024172 Cardiovascular disease Diseases 0.000 description 1
- 238000005481 NMR spectroscopy Methods 0.000 description 1
- 210000003484 anatomy Anatomy 0.000 description 1
- 239000008346 aqueous phase Substances 0.000 description 1
- 208000015322 bone marrow disease Diseases 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 208000029078 coronary artery disease Diseases 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 230000002526 effect on cardiovascular system Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000008595 infiltration Effects 0.000 description 1
- 238000001764 infiltration Methods 0.000 description 1
- 238000007918 intramuscular administration Methods 0.000 description 1
- 230000005865 ionizing radiation Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 208000030159 metabolic disease Diseases 0.000 description 1
- 210000003205 muscle Anatomy 0.000 description 1
- 238000001208 nuclear magnetic resonance pulse sequence Methods 0.000 description 1
- 206010033675 panniculitis Diseases 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000007306 turnover Effects 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Landscapes
- Apparatus For Radiation Diagnosis (AREA)
Abstract
The application relates to the field of intelligent medical treatment, in particular to an automatic fat quantifying method, system and equipment based on vertebral body segmentation. Comprises the steps of obtaining a medical image of a subject; dividing the vertebral body of the medical image to obtain a vertebral body; obtaining a body segment based on structural positioning of the vertebral body; a fat area and/or volume of a corresponding level of the body segment is calculated. The application can be used for automatically dividing various fat of the whole body, measuring the volume and measuring the fat area of each vertebral level, and has good clinical value.
Description
Technical Field
The application relates to the field of intelligent medical treatment, in particular to an automatic fat quantifying method, an automatic fat quantifying system, automatic fat quantifying equipment and a readable storage medium based on vertebral body segmentation.
Background
Fat content and distribution changes are associated with the occurrence and development of various diseases, and previous studies have confirmed that visceral fat in the ventral basin is a risk factor for various cardiovascular and metabolic diseases, that increased pericardial fat is associated with the presence and severity of coronary artery disease, that bone marrow fat fraction contributes to diagnosis of bone marrow diseases, etc.
The clinical noninvasive evaluation of fat content mainly adopts CT and MRI, wherein the magnetic resonance Dixon technology utilizes the chemical displacement difference of water and fat to obtain in-phase, anti-phase and water-fat separation images, and compared with CT, the method has the advantage of no ionizing radiation. The traditional method for measuring the fat content of each part such as subcutaneous fat, visceral fat at the abdomen and basin from the magnetic resonance Dixon image is to manually mark the fat area of each layer by doctor, multiply the fat area of each layer by layer thickness and add the layers layer by layer to obtain the fat volume. This procedure is time consuming and laborious and difficult to apply clinically, and thus in practice often the fat area of a particular vertebral level is measured, e.g. the commonly used L3 level visceral fat area reflects the abdominal visceral fat content.
In recent years, deep learning is widely applied in the field of medical image segmentation, and some methods have been tried on the problems of fat segmentation and quantification of magnetic resonance Dixon images, but the existing methods have some defects: (1) The existing method mainly focuses on the subcutaneous fat and visceral fat segmentation of the abdomen, and can not simultaneously segment and measure mediastinum fat, supraclavicular fat and bone marrow; (2) The existing method can not locate the anatomical position of the layer and can not provide the information of the fat area of the specific vertebral layer commonly used in clinic; (3) Since the fat region labeling process often relies on thresholds set according to the image gray scale, the deep-learning model that directly predicts fat regions in the image may deviate when migrating to out-of-domain data from different devices or scan parameters.
Disclosure of Invention
In order to solve the above problems, the present invention provides an automatic fat quantification method based on vertebral body segmentation, comprising:
acquiring a medical image of a subject;
dividing the vertebral body of the medical image to obtain a vertebral body;
Obtaining a body segment based on structural positioning of the vertebral body;
a fat area and/or volume of a corresponding level of the body segment is calculated.
Further, the positioning is to position the body segment from top to bottom by adopting a connected domain marking algorithm;
Preferably, the fat of the corresponding level of the body segment is divided into chest fat, abdomen fat;
preferably, the thoracic fat is fat from the upper edge of the 1 st thoracic vertebra to the upper edge portion of the 10 th thoracic vertebra;
Preferably, the abdominal fat is fat from the upper edge of the 10 th thoracic vertebra to the upper edge portion of the pelvis;
Preferably, the method further comprises segmenting the medical image to obtain pelvis and/or femur, calculating fat of corresponding layers of the body segment based on the body segment obtained by positioning the pelvis and/or femur, wherein the fat of the corresponding layers is pelvic fat and/or thigh fat;
preferably, the pelvic fat is the fat of the upper pelvic rim to lower pelvic rim portion;
Preferably, the thigh fat is the fat from the lower pelvic rim to the lower femoral rim portion.
The segmenting further comprises segmenting other fat regions of the medical image, the other fat regions comprising one or more of the following regions: subcutaneous fat region, neck-supraclavicular-axillary fat region, mediastinum region, abdominal pelvic visceral region.
Further, the segmentation process further comprises fine segmentation, and the part of the other fat region image, the gray value of which is smaller than a preset threshold value, is removed to obtain a fat image;
Preferably, the subcutaneous fat region, the neck-supraclavicular-axillary fat region, the mediastinum region and the abdominal viscera region in the other fat region are finely divided to obtain subcutaneous fat, neck-supraclavicular-axillary fat, mediastinum fat and abdominal viscera fat;
preferably, the preset threshold value is adaptively determined or set by using a gray level histogram.
The fat area is calculated by multiplying the real area of each pixel in the fat axis position image by the number of pixels;
Preferably, the fat area comprises one or more of the following: subcutaneous fat area, neck-supraclavicular-axillary fat area, mediastinal fat area, abdominal visceral fat area, bone marrow fat area, thigh fat area;
Preferably, the calculating of the area further comprises calculating a centrum central level fat area by first calculating a centrum central level ordia i; multiplying the real area of each pixel in the axial image of the ordinal number i layer by the number of pixels to calculate;
preferably, the calculation formula of the central layer sequence number of the vertebral body is:
Wherein the maximum level number on the vertical axis is i max, the minimum level number is i min, and Round (·) represents an integer.
The volume is calculated by multiplying the fat area of each layer by the layer thickness and adding the layers layer by layer;
preferably, the volume calculation formula is:
wherein the maximum layer number on the vertical axis is i max, and the minimum layer number is i min,SATi
Fat area, h is layer thickness;
Preferably, the volume comprises one or more of the following: subcutaneous fat volume, neck-supraclavicular-axillary fat volume, mediastinal fat volume, abdominal visceral fat volume, bone marrow fat volume, thigh fat volume;
preferably, the volume further comprises a thigh inter-muscular fat volume calculated by subtracting the thigh subcutaneous fat volume from the bone marrow fat volume.
The calculation further includes a fat fraction obtained by removing the signal intensity of the in-phase image from the signal intensity of the fat phase image;
Preferably, the fat fraction relates to a part comprising: vertebral body, pelvis, humerus, femur.
The method further comprises the step of averaging the fat area of the chest and the fat area of the abdomen to obtain the average fat area of the upper body; summing the chest fat volume and the abdomen fat volume to obtain an upper body fat volume;
optionally, the method further comprises averaging the mediastinum fat area and the visceral fat area of the abdominal basin to obtain an average fat area of the inner part of the upper body; summing the mediastinal fat volume and the abdominal visceral fat volume to obtain the internal fat volume of the upper body;
Optionally, the method further comprises averaging the chest fat area, the abdomen fat area and the basin fat area to obtain a trunk fat average area; summing the chest fat volume, the abdomen fat volume and the basin fat volume to obtain a trunk fat volume;
Optionally, the method further comprises averaging the chest fat area, the abdomen fat area, the basin fat area and the thigh fat area to obtain the average areas of the body fat and the lower limb fat; summing the chest fat volume, the abdomen fat volume, the basin fat volume and the thigh fat volume to obtain a trunk fat volume and a lower limb fat volume;
Optionally, the method further comprises averaging the chest fat area, the abdomen fat area, the basin fat area, the thigh fat area, the neck-supraclavicle-armpit fat area to obtain a whole body fat average area; the total body fat volume is obtained by summing the chest fat volume, the abdomen fat volume, the basin fat volume, the thigh fat volume, the neck-supraclavicle-armpit fat volume.
The invention aims to provide an automatic fat quantifying system based on vertebral body segmentation, which comprises the following components:
the acquisition module is used for: acquiring a medical image of a subject;
And a segmentation module: dividing the vertebral body of the medical image to obtain a vertebral body;
And a positioning module: obtaining a body segment based on structural positioning of the vertebral body;
The calculation module: a fat area and/or volume of a corresponding level of the body segment is calculated.
The invention aims to provide automatic fat quantifying equipment based on vertebral body segmentation, which comprises the following components:
A memory and a processor, the memory for storing program instructions; the processor is used for calling program instructions, and when the program instructions are executed, the automatic fat quantification method based on the vertebral body segmentation is realized.
An object of the present invention is to provide a computer-readable storage medium having stored thereon a computer program comprising:
The computer program, when executed by a processor, enables automatic quantification of fat based on segmentation of vertebral bodies as defined in any one of the claims.
The invention has the advantages that:
1. the method can output fat volume of multiple parts, fat fraction of each main bone and fat area of each layer of each vertebral body, comprehensively reflect the whole body fat distribution condition and create more possibility for deeply researching the association of fat distribution and diseases;
2. the method has good robustness and mobility, and can be well applied to patients with fat distribution change and data from different equipment or scanning parameters.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic flow chart of an automatic fat quantification method based on vertebral body segmentation according to an embodiment of the present invention;
Fig. 2 is a schematic diagram of an automatic fat quantification system based on vertebral body segmentation according to an embodiment of the present invention;
fig. 3 is a schematic diagram of an automatic fat quantification apparatus based on vertebral body segmentation according to an embodiment of the present invention;
fig. 4 is a schematic diagram of a whole body fat segmentation process according to an embodiment of the present invention;
Fig. 5 is a schematic diagram of a whole body fat segmentation result according to an embodiment of the present invention.
Detailed Description
In order to enable those skilled in the art to better understand the present invention, the following description will make clear and complete descriptions of the technical solutions according to the embodiments of the present invention with reference to the accompanying drawings.
In some of the flows described in the specification and claims of the present invention and in the above figures, a plurality of operations appearing in a particular order are included, but it should be clearly understood that the operations may be performed in other than the order in which they appear herein or in parallel, the sequence numbers of the operations such as S101, S102, etc. are merely used to distinguish between the various operations, and the sequence numbers themselves do not represent any order of execution. In addition, the flows may include more or fewer operations, and the operations may be performed sequentially or in parallel. It should be noted that, the descriptions of "first" and "second" herein are used to distinguish different messages, devices, modules, etc., and do not represent a sequence, and are not limited to the "first" and the "second" being different types.
Fig. 1 is a schematic diagram of an automatic fat quantification method based on vertebral body segmentation according to an embodiment of the present invention, which specifically includes:
s101: acquiring a medical image of a subject;
in one embodiment, the medical image employs one or more of the following: CT images and nuclear magnetic resonance images.
In one embodiment, in magnetic resonance imaging, the resonance frequencies of hydrogen protons in adipose tissue and hydrogen protons in other tissues in the human body are different due to their different molecular environments; when hydrogen protons of fat and other tissues are excited by the radio frequency pulse at the same time, their relaxation times are also different. Signals are acquired at different echo times, and adipose tissue and non-adipose tissue exhibit different signal intensities. With the above characteristics of different tissues in the human body, various pulse sequences for suppressing fat signals have been developed.
In one embodiment, the Dixon, magnetic resonance water-fat separation technique, is based on fat and water at different resonance frequencies. Dixon is based on TSE or 3DGRE sequences, which can generate four contrast images including positive phase (water and fat phase coincident), negative phase (water and fat phase opposite), and fat phase and water phase images generated by post-processing calculations.
S102: dividing the vertebral body of the medical image to obtain a vertebral body;
In one embodiment, the segmentation model employs one or more of the following: U-Net, faster R-
CNN、Deeplab V3、YOLACT、Mask R-CNN、RGA U-Net、PSPNet、SegNet、
FCN、SegGPT、SAM、SEEM。
In one embodiment, U-Net is a semantic segmentation algorithm for base depth learning. The method can enable a computer to automatically identify objects in different areas in an image, and meanwhile, each pixel can be classified, so that image segmentation at the pixel level is realized. The name of U-Net is derived from the shape of its network structure like U-shape, and its design purpose is to solve the problems existing in medical image segmentation.
The network structure of U-Net mainly consists of two parts, one is a contracted path and the other is an expanded path. The contracted path is used to extract feature information of the image, while the expanded path is used to achieve pixel-level segmentation. In the shrink path, the resolution of the input image is reduced from the original high resolution to the low resolution through some operations of rolling and pooling, and feature information is extracted. In the expansion path, the network restores the high-resolution feature mapping through deconvolution operation, so that a segmentation result with the same resolution as the original image is obtained.
A span connection is arranged between the contracted path and the expanded path of the U-Net, and the span connection is designed to keep high-quality characteristic information. After the features have undergone the convolution and pooling operations, their location information has been lost. If the deconvolution operation is directly performed at this time, inaccurate segmentation results may be caused. The use of span connections can avoid this problem by stitching feature maps at the same position between the contracted path and the expanded path, so that position information can be retained when deconvolution is performed, and the accuracy of segmentation is improved.
U-Net has some other features, such as the fact that the activation function it uses is a ReLU rather than a sigmoid or tanh, which can speed up the training process of the network. In addition, the U-Net also adopts data enhancement technology such as rotation, turnover and scaling to increase the diversity of the training set, thereby improving the robustness of the model.
In general, the principle of U-Net is based on a deep-learning convolutional neural network, which uses contracted and expanded paths to extract and recover feature information, while using span connections to preserve position information, thereby achieving high-precision semantic segmentation. The application range of the method is very wide, such as medical image segmentation, remote sensing image segmentation and the like.
In one embodiment, the segmenting further comprises segmenting other fat regions of the medical image, the other fat regions comprising one or more of the following regions: subcutaneous fat region, neck-supraclavicular-axillary fat region, mediastinum region, abdominal pelvic visceral region.
In one embodiment, the segmentation process further includes fine segmentation, and removing the part of the other fat region image with the gray value smaller than the preset threshold value to obtain a fat image;
Preferably, the subcutaneous fat region, the neck-supraclavicular-axillary fat region, the mediastinum region and the abdominal viscera region in the other fat region are finely divided to obtain subcutaneous fat, neck-supraclavicular-axillary fat, mediastinum fat and abdominal viscera fat;
preferably, the preset threshold value is adaptively determined or set by using a gray level histogram.
In one embodiment, the flow of whole body fat segmentation is shown in fig. 4, and the segmentation scheme is as follows:
(1) Establishing a whole-body magnetic resonance image dataset: one sample of the data set is that all Dixon aqueous phase, fat phase, same phase and opposite phase images are acquired in the same examination of a patient, the Dixon same phase images are used as references to register other images, the data set is divided into a training set, a verification set and a test set, and subcutaneous fat areas, neck-supraclavicle-armpit fat areas, mediastinum areas, abdominal pelvic visceral areas and vertebral bodies, pelvis, humerus and femur marrow are manually marked.
(2) Rough segmentation: constructing a deep learning model, training an image segmentation network by using all axial images in a training set, determining optimal super parameters of the training model according to the effect of a verification set, predicting four images of each axial layer of each sample on a test set, and adopting the category with the highest average prediction probability, thereby segmenting subcutaneous fat regions, neck-collarbone-armpit fat regions, mediastinum regions, abdominal viscera regions and bone marrow.
(3) Fine segmentation: and (3) according to the segmentation results of the subcutaneous fat region, the neck-collarbone-armpit fat region, the mediastinum region and the abdominal viscera region given by the model and the threshold value set by a user or adaptively determined based on a gray level histogram, respectively defining pixels with the Dixon fat phase image gray level larger than the threshold value as subcutaneous fat, neck-collarbone-armpit fat, mediastinum fat and abdominal viscera fat.
In one embodiment, a segmentation model is constructed based on a U-Net architecture, wherein the segmentation model consists of an input layer, N groups of downsampling modules, N groups of upsampling modules and an output layer, N is a natural number greater than 1, the downsampling modules consist of a convolution layer and a pooling layer, and the upsampling modules consist of transposed convolution. And after the whole body image is input through the input layer, carrying out feature extraction by an up-sampling module to obtain a feature image, recovering the size of the image through the up-sampling module, carrying out feature fusion on the feature images with different sizes obtained by each down-sampling module through jump connection and the features of the corresponding up-sampling module to obtain a final feature image, and inputting the final feature image into the output layer to obtain a segmentation result. The output layer adopts a convolution check of 1×1 to carry out linear transformation on the final feature map, generates class probability of each pixel, and finally converts the probability into class labels of each pixel by using a softmax activation function.
Further, the loss function of the segmentation model adopts the Dice loss, and the optimizer adopts Adam.
In one embodiment, the fat quantification method based on vertebral body segmentation of the present invention shows the whole body fat visualization results, and the Dice coefficients of each category are shown in table 1.
TABLE 1
S103: obtaining a body segment based on structural positioning of the vertebral body;
In one embodiment, the positioning is performed by using a connected domain labeling algorithm to position the body segment from top to bottom;
Preferably, the fat of the corresponding level of the body segment is divided into chest fat, abdomen fat;
preferably, the thoracic fat is fat from the upper edge of the 1 st thoracic vertebra to the upper edge portion of the 10 th thoracic vertebra;
Preferably, the abdominal fat is fat from the upper edge of the 10 th thoracic vertebra to the upper edge portion of the pelvis;
Preferably, the method further comprises segmenting the medical image to obtain pelvis and/or femur, calculating fat of corresponding layers of the body segment based on the body segment obtained by positioning the pelvis and/or femur, wherein the fat of the corresponding layers is pelvic fat and/or thigh fat;
preferably, the pelvic fat is the fat of the upper pelvic rim to lower pelvic rim portion;
Preferably, the thigh fat is the fat from the lower pelvic rim to the lower femoral rim portion.
In one embodiment, the portion of subcutaneous fat between the upper 1 st thoracic vertebra (inclusive) and the upper 10 th thoracic vertebra (exclusive) is defined as the thoracic subcutaneous fat volume, the portion of the upper 10 th thoracic vertebra (inclusive) and the upper pelvic edge (exclusive) is defined as the abdominal subcutaneous fat volume, the portion of the upper pelvic edge (inclusive) and the lower pelvic edge (inclusive) is defined as the pelvic subcutaneous fat volume, and the portion of the lower pelvic edge (exclusive) and the lower femoral edge (inclusive) is defined as the thigh subcutaneous fat volume.
S104: a fat area and/or volume of a corresponding level of the body segment is calculated.
In one embodiment, the fat area is calculated by multiplying the real area of each pixel in the fat axis image by the number of pixels;
Preferably, the fat area comprises one or more of the following: subcutaneous fat area, neck-supraclavicular-axillary fat area, mediastinal fat area, abdominal visceral fat area, bone marrow fat area, thigh fat area;
In one embodiment, the calculating of the area further comprises calculating a central level fat area of the vertebral body by first calculating a central level number i of the vertebral body; multiplying the real area of each pixel in the axial image of the ordinal number i layer by the number of pixels to calculate;
preferably, the calculation formula of the central layer sequence number of the vertebral body is:
Wherein the maximum level number on the vertical axis is i max, the minimum level number is i min, and Round (·) represents an integer.
In one embodiment, the volume is calculated by multiplying the fat area of each layer by the layer thickness and summing the layers one by one;
preferably, the volume calculation formula is:
wherein the maximum layer number on the vertical axis is i max, and the minimum layer number is i min,SATi
Fat area, h is layer thickness;
in one embodiment, the volume comprises one or more of the following: subcutaneous fat volume, neck-supraclavicular-axillary fat volume, mediastinal fat volume, abdominal visceral fat volume, bone marrow fat volume, thigh fat volume;
preferably, the volume further comprises a thigh inter-muscular fat volume calculated by subtracting the thigh subcutaneous fat volume from the bone marrow fat volume.
In one embodiment, the calculating further comprises a fat fraction obtained by removing signal intensities of the fat phase images from signal intensities of the in-phase images;
Preferably, the fat fraction relates to a part comprising: vertebral body, pelvis, humerus, femur.
In a specific embodiment, the subcutaneous fat segmentation results given for the model, the fat areas of the layers are multiplied by the layer thicknesses and added layer by layer to give the whole body subcutaneous fat volume.
For neck-supraclavicular-axillary fat, mediastinal fat, abdominal visceral fat, the fat areas of the layers were multiplied by the layer thickness and added layer by layer to give the corresponding fat volumes.
For bone marrow, dividing the gray level of the water phase image by the gray level of the same phase image to obtain fat fraction, and outputting average fat fraction and standard deviation of each vertebral body, pelvis, humerus and femur.
And calculating various fat areas of the layers of the centrum of each vertebra body to obtain the result.
In a specific embodiment, the calculation of fat area. In an axial image, let the number of pixels of a certain fat predicted by a model be x AT, and the real area corresponding to each pixel be S pixel, then the fat area S AT of the image is:
SAT＝xAT×Spixel (1)
Spixel＝x_spacing×y_spacing (2)
Wherein, x_spacing and y_spacing are image parameters of Dixon images. The subcutaneous fat (subcutaneous adipose tissue, SAT) area, visceral fat (visceral adipose tissue, VAT) area, bone marrow fat (bone marrow adipose tissue, BMAT) area, and the like can be calculated from formulas (1) to (2) as needed.
In one embodiment, the fat area of the central level of the vertebral body is calculated. Assuming that the maximum layer number of the model on the vertical axis of the segmentation result of a certain vertebral body is i max and the minimum layer number is i min, the central layer number i of the vertebral body is:
wherein Round (·) represents an integer. And (3) acquiring an axial position image of the layer i, and calculating the subcutaneous fat area and the visceral fat area of the central layer of the vertebral body according to formulas (1) - (2).
In one specific embodiment, the calculation of fat volume:
Chest, abdomen, basin: let the maximum layer number of the part to be measured (such as chest) on the vertical axis be i max, and the minimum layer number be i min. Assuming that the thickness of the layer of the Dixon image in the vertical axis direction is h, calculating the fat area S ATi of the category to be measured (such as subcutaneous fat) according to formulas (1) - (2) for each layer i with the ordinal number between i min～imax, and determining the fat volume V AT of the category to be measured (such as chest subcutaneous fat) of the portion to be measured as follows:
Thigh: thigh fat is mainly divided into three parts, namely thigh subcutaneous fat, bone marrow fat and intramuscular fat, and the thigh subcutaneous fat volume V thighSAT and the bone marrow fat volume V thighBMAT can be calculated respectively by referring to the formula (4). Inputting the thigh area into a fine segmentation module to obtain the total thigh fat, and calculating the total thigh fat volume V thighAT by referring to the formula (4). The volume of interthigh fat is:
VthighIMAT＝VthighAT-VthighSAT-VthighBMAT (5)
In one particular embodiment, the fat fraction is calculated: fat fraction: taking the vertebral body fat fraction as an example, a model is set to give a segmentation result to a certain vertebral body, n voxels are total, and the fat fraction FF i of any voxel i of the vertebral body is as follows:
Where FP i represents the signal intensity of voxel i in the fat phase (fatphase) image and IP i represents the signal intensity in the in-phase (inphase) image. Average value of the vertebral body fat fraction And standard deviation FF σ are:
In one embodiment, fat volume is measured as the size of the space occupied by adipose tissue and fat fraction is measured as the degree (percent) of fat infiltration in tissue organs (e.g., bone marrow, liver, spleen, kidney, pancreas).
In a specific embodiment, the main objective of the present invention is to evaluate the whole body fat distribution as comprehensively as possible from the Dixon images. Existing studies generally only evaluate the distribution of subcutaneous fat and visceral fat, but bone marrow fat is also an important component of body fat, and appears as one of the significantly high signal areas in the Dixon fat phase images, so that a model of simultaneous segmentation of bone marrow can achieve more comprehensive whole body fat segmentation and quantification, and reduce the risk of misrecognition of bone marrow fat as visceral fat.
In a specific embodiment, most of the fat quantitative studies currently only measure the fat area of a specific vertebral level due to the complexity of fat volume measurement, and the measurement of fat volume is mainly related to scientific research at present, no widely-applied normal reference value is established yet, and the differences of the selection of the vertebral level and the labeling of the fat area are large for doctors. Therefore, the objective quantitative fat area of the specific vertebral anatomy level is obtained based on the vertebral segmentation result, and the method has better compatibility with clinical practice.
By using a bone structure such as a vertebral body as an anatomical landmark, the chest (including the 1 st thoracic upper edge (inclusive) to the 10 th thoracic upper edge (exclusive)), the abdomen (including the 10 th thoracic upper edge (inclusive) to the pelvic upper edge (exclusive)), the basin (including the pelvic upper edge (inclusive) to the pelvic lower edge (inclusive)), and the thigh (including the pelvic lower edge (exclusive) to the femur lower edge (inclusive)) can be divided, and the total fat volume of each site can be calculated.
In one embodiment, body segments can be located through vertebral bodies, and then axial images of specific segments are intercepted, an organ segmentation model is applied, for example, abdominal organs (liver, pancreas, spleen and kidney) are segmented by using a layer between T6 and L3, and then organ fat fractions are calculated; heart segmentation was performed with a layer between T1 and T10, and epicardial fat volume and pericardial fat volume were measured.
In one embodiment, the method further comprises averaging the chest fat area, the abdomen fat area to obtain an average upper body fat area; summing the chest fat volume and the abdomen fat volume to obtain an upper body fat volume;
optionally, the method further comprises averaging the mediastinum fat area and the visceral fat area of the abdominal basin to obtain an average fat area of the inner part of the upper body; summing the mediastinal fat volume and the abdominal visceral fat volume to obtain the internal fat volume of the upper body;
Optionally, the method further comprises averaging the chest fat area, the abdomen fat area and the basin fat area to obtain a trunk fat average area; summing the chest fat volume, the abdomen fat volume and the basin fat volume to obtain a trunk fat volume;
Optionally, the method further comprises averaging the chest fat area, the abdomen fat area, the basin fat area and the thigh fat area to obtain the average areas of the body fat and the lower limb fat; summing the chest fat volume, the abdomen fat volume, the basin fat volume and the thigh fat volume to obtain a trunk fat volume and a lower limb fat volume;
Optionally, the method further comprises averaging the chest fat area, the abdomen fat area, the basin fat area, the thigh fat area, the neck-supraclavicle-armpit fat area to obtain a whole body fat average area; the total body fat volume is obtained by summing the chest fat volume, the abdomen fat volume, the basin fat volume, the thigh fat volume, the neck-supraclavicle-armpit fat volume.
In one embodiment, the thoracic fat comprises subcutaneous thoracic fat, mediastinal fat, vertebral bone marrow fat at the upper edge of the 1 st thoracic vertebra to the 10 th thoracic vertebra (not included); abdominal fat includes abdominal subcutaneous fat, abdominal visceral fat, 10 th thoracic vertebra upper edge (inclusive) -pelvic upper edge (exclusive) bone marrow fat; the pelvic fat comprises pelvic subcutaneous fat, pelvic visceral fat, pelvic upper edge (containing) -pelvic lower edge (containing) bone marrow fat; thigh fat includes subcutaneous thigh fat, interthigh muscle fat, pelvic lower edge (not included) -femoral lower edge (included) bone marrow fat.
In one embodiment, the calculation of fat area of the present invention includes one or more of the following:
1) Obtaining fat area calculations of body segments of the chest and/or abdomen based on the vertebral body segmentation localization;
2) Obtaining fat area calculation of the pelvic body segment based on pelvic segmentation positioning;
3) Obtaining fat area calculation of thigh body segments based on femur segmentation positioning;
4) Fat areas were calculated for other areas (subcutaneous fat area, neck-supraclavicular-axillary fat area, mediastinal fat area, abdominal visceral fat area, bone marrow fat area).
In one embodiment, the calculation of the fat volume of the present invention includes one or more of the following:
1) Obtaining fat volume calculations of body segments of the chest and/or abdomen based on the vertebral body segmentation localization;
2) Obtaining fat areas of body segments of the chest and/or abdomen based on the vertebral body segmentation positioning, and calculating fat volumes based on the fat areas;
3) Obtaining fat area calculation of the pelvic body segment based on pelvic segmentation positioning, and calculating fat volume based on the fat area;
4) Obtaining fat area calculation of thigh body segments based on femur segmentation positioning, and calculating fat volume based on the fat area;
5) Calculating fat areas of other regions (subcutaneous fat region, neck-supraclavicular-axillary fat region, mediastinal fat region, abdominal visceral fat region, bone marrow fat region), and calculating fat volume based on the fat areas;
6) Obtaining fat volume calculation of the pelvic body segment based on the pelvic segmentation localization;
7) Obtaining fat volume calculation of thigh body segments based on femur segmentation localization;
8) For other areas (subcutaneous fat area, neck-supraclavicular-axillary fat area, mediastinum fat)
Fat region, abdominal visceral fat region, bone marrow fat region). In one embodiment, the fat area/volume summation of the present invention comprises one or more of the following: 1) Averaging/summing the fat area/volume of the chest body segment with the fat area/volume of the abdomen body segment;
2) Averaging/summing the fat area/volume of the chest body segment and the fat area/volume of the basin body segment;
3) Averaging/summing the fat area/volume of the chest body segment and the fat area/volume of the thigh body segment;
4) Averaging/summing the fat area/volume of the abdominal body segment and the fat area/volume of the pelvic body segment;
5) Averaging/summing the fat area/volume of the abdominal body segment and the fat area/volume of the thigh body segment;
6) Averaging/summing the fat area/volume of the pelvic body segment with the fat area/volume of the thigh body segment;
7) Averaging/summing the fat area/volume of the chest body segment with the fat area/volume of the abdomen body segment with the fat area/volume of the basin body segment;
8) Averaging/summing the fat area/volume of the chest body segment with the fat area/volume of the abdomen body segment with the fat area/volume of the thigh body segment;
9) Averaging/summing the fat area/volume of the abdominal body segment with the fat area/volume of the pelvic body segment with the fat area/volume of the thigh body segment;
10 Averaging of fat area/volume of chest fat and neck-supraclavicular-axillary fat +.
Summing; averaging/summing the fat area/volume of abdominal fat and neck-supraclavicular-axillary fat; averaging/summing the fat area/volume of pelvic fat and neck-supraclavicular-axillary fat; averaging/summing the fat area/volume of thigh fat and neck-supraclavicular-axillary fat; averaging/summing of fat areas/volumes of chest fat, abdominal fat, neck-supraclavicular-axillary fat; averaging/summing of fat areas/volumes of chest fat, abdominal fat, pelvic fat, neck-supraclavicular-axillary fat; averaging/summing of fat areas/volumes of chest fat, abdominal fat, pelvic fat, thigh fat, neck-supraclavicular-axillary fat; fat area of visceral fat and mediastinal fat in the ventral basin +.
Averaging/summing of volumes; averaging/summing fat areas/volumes of chest fat, abdominal visceral fat, mediastinal fat; averaging/summing fat areas/volumes of chest fat, abdominal fat, pelvic fat, abdominal visceral fat and mediastinum fat; averaging/summing fat areas/volumes of chest fat, abdomen fat, pelvic fat, thigh fat, abdominal pelvic visceral fat and mediastinum fat; the fat areas/volumes of chest fat, abdomen fat, basin fat, thigh fat, neck-supraclavicle-axilla fat, abdomen basin visceral fat and mediastinum fat are averaged/summed.
In one embodiment, by inputting the subject medical image into the method of the invention, the fat area, fat volume and fat fraction are obtained, and the fat area, fat volume and fat fraction can quantify the whole body fat or a certain section of the whole body or a certain type of fat of the whole body, so that the research on diseases related to the fat can be carried out, the research on the internal connection of the whole body fat of the human body can be promoted, and the research possibility between diseases related to fat of different parts can be improved. In addition, the vertebral body segmentation result obtains objective quantitative fat area of the specific vertebral body anatomical layer, and has better compatibility with clinical practice.
In one embodiment, the method for quantifying whole body fat according to the present invention is based on obtaining fat volumes and/or areas of different body segments after vertebral body segmentation, such as obtaining fat volumes and/or areas of an abdominal segment, wherein the fat of the abdominal segment further comprises subcutaneous fat of the abdomen, visceral fat of the abdomen, bone marrow fat of the vertebral body of the abdominal segment.
In one embodiment, the method of the present invention may obtain fat volumes and/or areas of multiple body segments simultaneously, such as obtaining fat volumes and/or areas of an abdominal segment and a pelvic segment.
Fig. 2 is a schematic diagram of an automatic fat quantification system based on vertebral body segmentation according to an embodiment of the present invention, which specifically includes:
the acquisition module is used for: acquiring a medical image of a subject;
And a segmentation module: dividing the vertebral body of the medical image to obtain a vertebral body;
And a positioning module: obtaining a body segment based on structural positioning of the vertebral body;
The calculation module: a fat area and/or volume of a corresponding level of the body segment is calculated.
Fig. 3 is a schematic diagram of an automatic fat quantifying device based on vertebral body segmentation according to an embodiment of the present invention, which specifically includes:
A memory and a processor; the memory is used for storing program instructions; the processor is used for calling program instructions, and when the program instructions are executed, any one of the automatic fat quantification method based on vertebral body segmentation is executed.
A computer readable storage medium storing a computer program which, when executed by a processor, is any of the above-described automatic fat quantification methods based on vertebral body segmentation.
The results of the verification of the present verification embodiment show that assigning an inherent weight to an indication may improve the performance of the method relative to the default setting. It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, which are not repeated herein. In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of the units is merely a logical function division, and there may be additional divisions when actually implemented, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form. The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment. In addition, each functional unit in the embodiments of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units. Those of ordinary skill in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be implemented by a program to instruct related hardware, the program may be stored in a computer readable storage medium, and the storage medium may include: read Only Memory (ROM), random access Memory (RAM, random Access Memory), magnetic or optical disk, and the like.
It will be appreciated by those skilled in the art that all or part of the steps in the method of the above embodiment may be implemented by a program to instruct related hardware, where the program may be stored in a computer readable storage medium, where the medium may be a rom, a magnetic disk, or an optical disk, etc.
While the foregoing describes a computer device provided by the present invention in detail, those skilled in the art will appreciate that the foregoing description is not meant to limit the invention thereto, as long as the scope of the invention is defined by the claims appended hereto.
Claims (10)
1. An automatic fat quantification method based on vertebral body segmentation, which is characterized by comprising the following steps:
acquiring a medical image of a subject;
dividing the vertebral body of the medical image to obtain a vertebral body;
Obtaining a body segment based on structural positioning of the vertebral body;
a fat area and/or volume of a corresponding level of the body segment is calculated.
2. The automatic fat quantification method based on vertebral body segmentation according to claim 1, wherein the positioning is body segment positioning from top to bottom by adopting a connected domain labeling algorithm;
preferably, the fat of the corresponding level of the body segment is chest fat and/or abdomen fat;
preferably, the thoracic fat is fat from the upper edge of the 1 st thoracic vertebra to the upper edge portion of the 10 th thoracic vertebra;
Preferably, the abdominal fat is fat from the upper edge of the 10 th thoracic vertebra to the upper edge portion of the pelvis;
Preferably, the method further comprises segmenting the medical image to obtain pelvis and/or femur, calculating fat of corresponding layers of the body segment based on the body segment obtained by positioning the pelvis and/or femur, wherein the fat of the corresponding layers is pelvic fat and/or thigh fat;
preferably, the pelvic fat is the fat of the upper pelvic rim to lower pelvic rim portion;
Preferably, the thigh fat is the fat from the lower pelvic rim to the lower femoral rim portion.
3. The automatic fat quantification method based on vertebral body segmentation according to claim 1, wherein the segmentation further comprises segmentation of other fat regions of the medical image, the other fat regions comprising one or more of the following regions: subcutaneous fat region, neck-supraclavicular-axillary fat region, mediastinum region, abdominal pelvic visceral region.
Preferably, the segmentation process further comprises fine segmentation, and the part of the other fat region image, the gray value of which is smaller than a preset threshold value, is removed to obtain a fat image;
Preferably, the subcutaneous fat region, the neck-supraclavicular-axillary fat region, the mediastinum region and the abdominal viscera region in the other fat region are finely divided to obtain subcutaneous fat, neck-supraclavicular-axillary fat, mediastinum fat and abdominal viscera fat;
preferably, the preset threshold value is adaptively determined or set by using a gray level histogram.
4. The automatic fat quantification method based on vertebral body segmentation according to any one of claims 1 to 3, wherein the fat area is calculated by multiplying the real area of each pixel in the fat axis image by the number of pixels;
Preferably, the calculating of the area further comprises calculating a centrum central level fat area by first calculating a centrum central level ordia i; multiplying the real area of each pixel in the axial image of the ordinal number i layer by the number of pixels to calculate;
preferably, the calculation formula of the central layer sequence number of the vertebral body is:
Wherein the maximum level number on the vertical axis is i max, the minimum level number is i min, and Round (·) represents an integer.
5. The automatic quantification method of fat based on vertebral body segmentation according to claim 1 or 4, wherein the volume is calculated by multiplying the fat area of each layer by the layer thickness and adding up the layers layer by layer;
preferably, the volume calculation formula is:
Wherein, the maximum layer number on the vertical axis is i max, the minimum layer number is i min,SATi, the fat area and h the layer thickness;
Preferably, the volume comprises one or more of the following: subcutaneous fat volume, neck-supraclavicular-axillary fat volume, mediastinal fat volume, abdominal visceral fat volume, bone marrow fat volume, thigh fat volume;
preferably, the volume further comprises a thigh inter-muscular fat volume calculated by subtracting the thigh subcutaneous fat volume from the bone marrow fat volume.
6. The automatic quantification of fat based on vertebral body segmentation according to claim 1 or 2, wherein the calculation further comprises calculation of a fat fraction obtained by removing the signal intensity of the in-phase image from the signal intensity of the fat phase image;
Preferably, the fat fraction relates to a part comprising: vertebral body, pelvis, humerus, femur.
7. The automatic fat quantification method based on vertebral body segmentation according to any one of claims 1 to 5, further comprising averaging the chest fat area, the abdomen fat area to obtain an average upper body fat area; summing the chest fat volume and the abdomen fat volume to obtain an upper body fat volume;
optionally, the method further comprises averaging the mediastinum fat area and the visceral fat area of the abdominal basin to obtain an average fat area of the inner part of the upper body; summing the mediastinal fat volume and the abdominal visceral fat volume to obtain the internal fat volume of the upper body;
Optionally, the method further comprises averaging the chest fat area, the abdomen fat area and the basin fat area to obtain a trunk fat average area; summing the chest fat volume, the abdomen fat volume and the basin fat volume to obtain a trunk fat volume;
Optionally, the method further comprises averaging the chest fat area, the abdomen fat area, the basin fat area and the thigh fat area to obtain the average areas of the body fat and the lower limb fat; summing the chest fat volume, the abdomen fat volume, the basin fat volume and the thigh fat volume to obtain a trunk fat volume and a lower limb fat volume;
Optionally, the method further comprises averaging the chest fat area, the abdomen fat area, the basin fat area, the thigh fat area, the neck-supraclavicle-armpit fat area to obtain a whole body fat average area; the total body fat volume is obtained by summing the chest fat volume, the abdomen fat volume, the basin fat volume, the thigh fat volume, the neck-supraclavicle-armpit fat volume.
8. An automatic fat quantification system based on vertebral body segmentation, comprising:
the acquisition module is used for: acquiring a medical image of a subject;
And a segmentation module: dividing the vertebral body of the medical image to obtain a vertebral body;
And a positioning module: obtaining a body segment based on structural positioning of the vertebral body;
The calculation module: a fat area and/or volume of a corresponding level of the body segment is calculated.
9. An automatic fat quantification apparatus based on vertebral body segmentation, comprising:
A memory and a processor, the memory for storing program instructions; the processor is configured to invoke program instructions, which when executed, implement the automatic fat quantification method based on vertebral body segmentation of any of the claims 1-7.
10. A computer-readable storage medium having a computer program stored thereon, comprising:
the computer program, when executed by a processor, implements the automatic quantification of fat based on segmentation of vertebral bodies according to any of the claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410191127.XA CN117994322A (en) | 2024-02-21 | 2024-02-21 | Automatic fat quantification method, system and equipment based on vertebral body segmentation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410191127.XA CN117994322A (en) | 2024-02-21 | 2024-02-21 | Automatic fat quantification method, system and equipment based on vertebral body segmentation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117994322A true CN117994322A (en) | 2024-05-07 |
Family
ID=90898920
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410191127.XA Pending CN117994322A (en) | 2024-02-21 | 2024-02-21 | Automatic fat quantification method, system and equipment based on vertebral body segmentation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117994322A (en) |
-
2024
- 2024-02-21 CN CN202410191127.XA patent/CN117994322A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US6799066B2 (en) | Technique for manipulating medical images | |
| EP1941453B1 (en) | Method of registering images, algorithm for carrying out the method of registering images, a program for registering images using the said algorithm and a method of treating biomedical images to reduce imaging artefacts caused by object movement | |
| Ortiz et al. | Automatic atlas‐based segmentation of the breast in MRI for 3D breast volume computation | |
| US10083506B2 (en) | Whole body image registration method and method for analyzing images thereof | |
| Datta et al. | Gray matter segmentation of the spinal cord with active contours in MR images | |
| CN101133431A (en) | Method and computer program product for registering biomedical images with reduced imaging arefacts caused by object movement | |
| KR20150045885A (en) | Systems and methods for registration of ultrasound and ct images | |
| US8417005B1 (en) | Method for automatic three-dimensional segmentation of magnetic resonance images | |
| Wald et al. | Spatial autocorrelation and mean intercept length analysis of trabecular bone anisotropy applied to in vivo magnetic resonance imaging | |
| CN110163877A (en) | A kind of method and system of MRI ventricular structure segmentation | |
| Sinha et al. | Exploration of male urethral sphincter complex using diffusion tensor imaging (DTI)‐based fiber‐tracking | |
| Wong et al. | A comparison of peripheral imaging technologies for bone and muscle quantification: a review of segmentation techniques | |
| US10136818B2 (en) | High resolution intraoperative MRI images | |
| CN115830016A (en) | Medical image registration model training method and equipment | |
| Yin et al. | Automatic breast tissue segmentation in MRIs with morphology snake and deep denoiser training via extended Stein’s unbiased risk estimator | |
| CN111429404B (en) | Imaging system and method for detecting cardiovascular and cerebrovascular vessels | |
| Reinertsen et al. | Vessel driven correction of brain shift | |
| CN112508880A (en) | Intracranial blood vessel image registration method, electronic device, and computer-readable storage medium | |
| CN117994322A (en) | Automatic fat quantification method, system and equipment based on vertebral body segmentation | |
| CN109741439A (en) | A kind of three-dimensional rebuilding method of two dimension MRI fetus image | |
| CN114821049A (en) | Method, device and equipment for segmenting image and storage medium | |
| Moreno et al. | Non linear transformation field to build moving meshes for patient specific blood flow simulations | |
| CN112508872A (en) | Intracranial blood vessel image preprocessing method and electronic equipment | |
| CN112819789A (en) | Apparatus and method for cardiac assessment | |
| CN112508878A (en) | Cerebrovascular image registration and fusion method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination |