CN111681254A - Catheter detection method and system for vascular aneurysm interventional operation navigation system - Google Patents
Catheter detection method and system for vascular aneurysm interventional operation navigation system Download PDFInfo
- Publication number
- CN111681254A CN111681254A CN202010546190.2A CN202010546190A CN111681254A CN 111681254 A CN111681254 A CN 111681254A CN 202010546190 A CN202010546190 A CN 202010546190A CN 111681254 A CN111681254 A CN 111681254A
- Authority
- CN
- China
- Prior art keywords
- catheter
- coding
- decoding
- deep learning
- sequence
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 69
- 206010002329 Aneurysm Diseases 0.000 title claims abstract description 46
- 238000000034 method Methods 0.000 claims abstract description 61
- 230000011218 segmentation Effects 0.000 claims abstract description 40
- 230000005540 biological transmission Effects 0.000 claims abstract description 11
- 238000013135 deep learning Methods 0.000 claims description 54
- 208000037170 Delayed Emergence from Anesthesia Diseases 0.000 claims description 45
- 125000004122 cyclic group Chemical group 0.000 claims description 38
- 238000001356 surgical procedure Methods 0.000 claims description 19
- 230000008569 process Effects 0.000 claims description 13
- 238000013152 interventional procedure Methods 0.000 claims description 12
- 230000006870 function Effects 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 9
- 230000000306 recurrent effect Effects 0.000 claims description 9
- 238000012549 training Methods 0.000 claims description 9
- 238000000605 extraction Methods 0.000 claims description 6
- 238000005070 sampling Methods 0.000 claims description 6
- 230000001537 neural effect Effects 0.000 claims description 5
- 238000013528 artificial neural network Methods 0.000 claims description 4
- 238000012795 verification Methods 0.000 claims description 4
- 101001013832 Homo sapiens Mitochondrial peptide methionine sulfoxide reductase Proteins 0.000 claims description 3
- 102100031767 Mitochondrial peptide methionine sulfoxide reductase Human genes 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 229920006395 saturated elastomer Polymers 0.000 claims description 3
- 230000008439 repair process Effects 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 7
- 230000033001 locomotion Effects 0.000 description 5
- 238000011282 treatment Methods 0.000 description 4
- 238000002583 angiography Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 2
- 210000004204 blood vessel Anatomy 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 238000002594 fluoroscopy Methods 0.000 description 2
- 238000011084 recovery Methods 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 230000002792 vascular Effects 0.000 description 2
- 208000024172 Cardiovascular disease Diseases 0.000 description 1
- LLQPHQFNMLZJMP-UHFFFAOYSA-N Fentrazamide Chemical compound N1=NN(C=2C(=CC=CC=2)Cl)C(=O)N1C(=O)N(CC)C1CCCCC1 LLQPHQFNMLZJMP-UHFFFAOYSA-N 0.000 description 1
- RTAQQCXQSZGOHL-UHFFFAOYSA-N Titanium Chemical compound [Ti] RTAQQCXQSZGOHL-UHFFFAOYSA-N 0.000 description 1
- 206010052428 Wound Diseases 0.000 description 1
- 208000027418 Wounds and injury Diseases 0.000 description 1
- 238000002679 ablation Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 208000007474 aortic aneurysm Diseases 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000002872 contrast media Substances 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
- 238000002059 diagnostic imaging Methods 0.000 description 1
- 230000008034 disappearance Effects 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 210000001105 femoral artery Anatomy 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000012010 growth Effects 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 230000004807 localization Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 239000003550 marker Substances 0.000 description 1
- 230000036285 pathological change Effects 0.000 description 1
- 231100000915 pathological change Toxicity 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 210000002321 radial artery Anatomy 0.000 description 1
- 238000010845 search algorithm Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 238000002560 therapeutic procedure Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
- 230000017105 transposition Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- 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
- G06N3/044—Recurrent networks, e.g. Hopfield 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
- G06N3/04—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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/194—Segmentation; Edge detection involving foreground-background segmentation
-
- 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
- G06T2207/10016—Video; Image sequence
-
- 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
- G06T2207/30101—Blood vessel; Artery; Vein; Vascular
Abstract
The invention belongs to the field, particularly relates to a catheter detection method for a vascular aneurysm interventional operation navigation system, and aims to solve the problem that a catheter cannot be inserted in an X-ray transmission image in an endovascular aneurysm repair operation in real time and accurately segmented and tracked. The invention comprises the following steps: the method comprises the steps of obtaining an X-ray transmission video sequence of a region containing a catheter as a sequence to be detected, generating a segmentation mask sequence through a trained coding and decoding structure according to the sequence to be detected, and covering a binary segmentation mask sequence on a video to be detected to obtain the video sequence of the catheter. The invention improves the accuracy and speed of the segmentation and tracking of the X-ray transmission image interventional catheter in the endovascular aneurysm repair operation, and meets the real-time performance required by the operation tracking algorithm.
Description
Technical Field
The invention belongs to the technical field of image processing, and particularly relates to a catheter detection method and system for a vascular aneurysm interventional operation navigation system.
Background
The minimally invasive interventional operation navigation system is a new emerging cardiovascular disease treatment system, namely a new method integrating multiple technologies such as computer science, artificial intelligence, automatic control, image processing, multi-mode fusion, target segmentation, three-dimensional imaging, virtual reality training and clinical treatment. The method uses medical images of various modes to assist doctors to puncture interventional operation instruments from the radial artery or the femoral artery and send the interventional operation instruments to the position with narrow blood vessels for treatment, and the method can improve operation quality, reduce operation wounds and reduce pain of patients.
Although this approach has made good clinical progress, minimally invasive interventional procedures have faced various difficulties in practical applications of aortic aneurysm surgery. For example, currently there is no perfect surgical planning method for certain intervention-related treatments; the doctor is difficult to accurately send the interventional therapy apparatus to the pathological change position according to the preset operation scheme, and the operation is implemented, and the catheter detection in the vascular interventional operation is an important part for realizing the vascular navigation system, but has the following difficulties: (1) the signal-to-noise ratio (SNR) of the X-ray image is low and background noise greatly interferes with the segmentation of the catheter. (2) The extreme foreground-background class imbalance is caused by the low ratio of conduit pixels to background pixels. (3) Due to contrast agents and line-like structures such as vertebrae and pelvic contours, edge pixels of the catheter become an example of misclassification.
With the development of the technology, the deep learning has great advantages in image detection and recognition, and detection accuracy and detection speed. Since convolution has great advantages in terms of images, including great advantages in terms of computation speed, learning ability.
The method for detecting the navigation instrument for the vascular aneurysm interventional operation is further deepened and improved on the basis of the patents, and obviously improves the detection speed, the detection precision, the robustness caused by individual difference and the like compared with the common instrument detection method.
Disclosure of Invention
In order to solve the above problems in the prior art, namely the problem that the catheter inserted in an X-ray transmission image in the endovascular aneurysm repair operation cannot be accurately segmented and tracked in real time, the invention provides a catheter detection method for a vascular aneurysm interventional operation navigation system, wherein the detection method comprises the following steps
Step S10, acquiring an X-ray transmission video sequence of an area containing a catheter in the operation process as a video sequence to be detected;
step S20, based on the video sequence to be detected, generating a binary segmentation mask sequence of the catheter through a trained coding and decoding structure based on deep learning;
step S30, covering the binary segmentation mask sequence on the video sequence to be detected to obtain a video sequence of the catheter;
the coding and decoding structure based on deep learning comprises a first convolutional layer, a first cyclic residual block, a multi-level nested coding and decoding structure and a second convolutional layer which are connected in sequence; the multi-level nested coding and decoding structure is formed by nesting and inserting a next-level coding and decoding structure between a coder and a decoder of each level of coding and decoding structure;
the encoder and the decoder of the multi-level nested coding and decoding structure respectively comprise a plurality of coding modules and a plurality of decoding modules; the coding module is connected with the corresponding peer decoding module through residual connection;
the encoding module is constructed based on a MobileNet V2 network and comprises an encoding residual block and a second cyclic residual block which are connected in sequence;
the decoding module is constructed based on a U-net neural structure and a recurrent neural network and comprises a decoding block and a second cyclic residual block which are connected in sequence.
Further, step S20 includes:
step S21, generating a first characteristic image for any picture to be detected in the video sequence to be detected through a first convolution layer in the coding and decoding structure based on the deep learning;
step S22, generating a second feature image from the first feature image through a first cyclic residual block in the coding and decoding structure based on the deep learning;
step S23, based on the second feature image, performing hierarchical coding through each coding module in the coding and decoding structure based on the deep learning to obtain a feature compressed image;
step S24, based on the feature compressed image, through each decoding module in the coding and decoding structure based on the deep learning, the hierarchical decoding is carried out, and the up-sampling feature image is obtained;
step S25, based on the up-sampling feature image, generating a binary segmentation mask of the catheter corresponding to the picture to be detected through the second convolution layer in the coding and decoding structure based on the deep learning, and obtaining a binary segmentation mask sequence of the catheter.
Further, the second cyclic residual block is a residual block including two cyclic convolution layers.
Further, the encoded residual block is a residual block including a fourth convolutional layer, a Dwise convolutional layer, and a fifth convolutional layer.
Further, the decoding block includes a sixth convolutional layer, a transposed convolutional layer, and a seventh convolutional layer, which are connected in sequence.
Further, the sixth convolutional layer and the seventh convolutional layer are convolutional layers incorporating a batch regularization layer.
Further, the deep learning based codec structure adopts a loss function in training as follows:
where LBCE is a binary cross-entropy loss function, β i is a gradient density coordination parameter,
is the pixel label of the ith pixel, 1 is the catheter, 0 is the background, pi is
And finally, predicting the value.
Further, the deep learning-based coding and decoding structure is initialized by adopting a MSRA method.
Furthermore, the deep learning-based coding and decoding structure reduces the learning rate by 2 times when the verification precision is saturated through multiple iterations in the training.
In another aspect of the present invention, a catheter detection system for a vascular aneurysm interventional surgery navigation system is provided, the detection system comprising: a video acquisition unit 100, a mask generation unit 200, and a result generation unit 300;
a video acquisition unit 100, configured to acquire an X-ray transmission video sequence of a region including a catheter during a surgical procedure as a video sequence to be detected;
a mask generating unit 200, which generates a binary segmentation mask sequence of the catheter based on the video sequence to be detected through a trained coding and decoding structure based on deep learning;
and a result generating unit 300, which overlays the binary segmentation mask sequence on the video sequence to be detected to obtain a video sequence of the catheter.
Further, the mask generating unit 200 includes:
a video extraction subunit 210, configured to generate, for any to-be-detected picture in the to-be-detected video sequence, a first feature image through a first convolution layer in the deep learning-based coding and decoding structure;
an image preprocessing subunit 220, configured to generate a second feature image from the first feature image through a first cyclic residual block in a deep learning-based codec structure;
an image encoding subunit 230, configured to perform hierarchical encoding through each encoding module in the deep learning-based encoding and decoding structure based on the second feature image, to obtain a feature compressed image;
an image decoding subunit 240, configured to perform hierarchical decoding on the feature compressed image through each decoding module in the deep learning-based coding and decoding structure, so as to obtain an upsampled feature image;
a mask generating subunit 250, configured to generate, based on the upsampled feature image, a binary segmentation mask of the catheter corresponding to the picture to be detected through the second convolutional layer in the deep learning-based coding and decoding structure, so as to obtain a binary segmentation mask sequence of the catheter.
In a third aspect of the present invention, a storage device is provided, in which a plurality of programs are stored, the programs being adapted to be loaded and executed by a processor to implement the above-mentioned catheter detection method for a vessel aneurysm interventional surgery navigation system.
In a fourth aspect of the present invention, a processing apparatus is provided, which includes a processor, a storage device; the processor is suitable for executing various programs; the storage device is suitable for storing a plurality of programs; the program is suitable to be loaded and executed by a processor to implement the above-mentioned catheter detection method for a vascular aneurysm interventional surgery navigation system.
The invention has the beneficial effects that:
(1) according to the catheter detection method for the vascular aneurysm interventional operation navigation system, the feature extraction structure in the U-net structure is replaced by the encoding residual block invoked Res-blocks which occupies less computing resources and has higher computing speed, so that the catheter detection speed in an operation image is improved, and the real-time performance required by a tracking algorithm of an operation is met.
(2) The invention discloses a catheter detection method for a vascular aneurysm interventional operation navigation system, which combines a cyclic Residual block Current Residual constant.Block with an encoding Residual block invoked Res-blocks, extracts more context characteristic information in a characteristic extraction structure, and improves the accuracy and speed of catheter detection in an operation image.
(3) The method for detecting the catheter for the vascular aneurysm interventional operation navigation system improves the accuracy of image detail recovery by combining the Recurrent residual conv.Block with the decoding fast Decoder Block in the U-net neural structure, improves the accuracy of catheter detection in an operation image by matching with good convolution of a coding module, and obtains accurate pixel-level prediction and good segmentation results.
(4) The invention relates to a catheter detection method for a vascular aneurysm interventional operation navigation system, which adopts a classification loss function
The method solves the problem of unbalanced target and background pixels and solves the problem that the edge of the sample cannot be accurately segmented.
Drawings
Other features, objects and advantages of the present application will become more apparent upon reading of the following detailed description of non-limiting embodiments thereof, made with reference to the accompanying drawings in which:
FIG. 1 is a schematic flow chart of a catheter detection method for a vascular aneurysm interventional procedure navigation system according to the present invention;
FIG. 2 is a schematic diagram of a deep learning-based encoding and decoding structure in a catheter detection method for a vascular aneurysm interventional surgery navigation system according to the present invention;
FIG. 3 is a schematic structural diagram of an encoded residual error block invoked Res-blocks in the catheter detection method for a vascular aneurysm interventional surgery navigation system according to the present invention;
FIG. 4 is a structural diagram of a circulating Residual error block Recurrent Residual Conv.Block in the catheter detection method for the vascular aneurysm interventional operation navigation system of the invention;
FIG. 5 is a schematic structural diagram of a Decoder Block in the catheter detection method for a vascular aneurysm interventional operation navigation system according to the invention;
FIG. 6 is an image of a catheter obtained by manual marking in an embodiment of the present invention;
FIG. 7 is an image of a catheter obtained using the U-net method of the prior art;
FIG. 8 is an image of a catheter obtained by the RCL method of the prior art;
FIG. 9 is a catheter image obtained by a method incorporating a fast codec structure of a secure Residual according to an embodiment of the present invention.
In fig. 2, the layer 1 is a convolutional layer Conv. + ReLU, the layer 2 is a cyclic convolutional Block recurturyalconv.block, the layers 4, 6, 8 and 10 are cyclic convolutional blocks recurturua conv.block in downsampling, the layers 12, 14, 16 and 18 are cyclic convolutional blocks recurup-conv.block in upsampling, the layers 3, 5, 7 and 9 are coded residual blocks inversed Res-blocks, the layers 11, 13, 15 and 17 are decoding blocks Decoder Block, the layer 19 is a convolutional layer Conv. + ReLU, and the layers a-H, B-G, C-F and D-E are corresponding coding and decoding structures connected by residual.
Detailed Description
The present application will be described in further detail with reference to the following drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the relevant invention and not restrictive of the invention. It should be noted that, for convenience of description, only the portions related to the related invention are shown in the drawings.
It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict. The present application will be described in detail below with reference to the embodiments with reference to the attached drawings.
The invention provides a catheter detection method for a vascular aneurysm interventional operation navigation system, which comprises the following steps:
step S10, acquiring an X-ray transmission video sequence of an area containing a catheter in the operation process as a video sequence to be detected;
step S20, based on the video sequence to be detected, generating a binary segmentation mask sequence of the catheter through a trained coding and decoding structure based on deep learning;
step S30, covering the known segmentation mask on the video sequence to be detected to obtain a video sequence of the catheter;
the coding and decoding structure based on deep learning comprises a first convolutional layer, a first cyclic residual block, a multi-level nested coding and decoding structure and a second convolutional layer which are connected in sequence; the multi-level nested coding and decoding structure is formed by nesting and inserting a next-level coding and decoding structure between a coder and a decoder of each level of coding and decoding structure;
the encoder and the decoder of the multi-level nested coding and decoding structure respectively comprise a plurality of coding modules and a plurality of decoding modules; the coding module is connected with the corresponding peer decoding module through residual connection;
the coding module is constructed based on a MobileNet V2 network and sequentially comprises a coding residual block and a second cyclic residual block;
the decoding module is sequentially a decoding block and a third cyclic residual block based on a U-net neural structure and a recurrent neural network component.
In order to more clearly describe the method for detecting a catheter used in a navigation system for vascular aneurysm interventional surgery according to the present invention, the following will describe each step in the method embodiment of the present invention in detail with reference to fig. 1, which is a schematic flow chart of the method for detecting a catheter used in a navigation system for vascular aneurysm interventional surgery according to the present invention.
Fig. 1 shows a flow diagram of a catheter detection method for a navigation system for an endovascular aneurysm interventional operation according to an embodiment of the present invention, which includes steps S10-S30, and the steps are described in detail as follows:
step S10, acquiring an X-ray transmission video sequence of an area containing a catheter in the operation process as a video sequence to be detected;
in this embodiment, a real-time video sequence in the whole operation process is extracted from an X-ray video sequence in the minimally invasive interventional operation process, so as to obtain a video sequence with an interventional operation catheter for data calibration. A new data set CTRSeg based on a two-dimensional X-ray fluoroscopy. Innova 3100-IQ digital flat-panel angiography (GE Healthcare) was used to obtain the clinical sequence of the most representative 30 patients. In detail, the data set was trained with 115 sequences from 20 patients (778 images), tested with 35 sequences from 6 patients (234 images) and validated with 25 sequences from 4 patients (156 images).
Step S20, based on the video sequence to be detected, generating a binary segmentation mask sequence of the catheter through a trained coding and decoding structure based on deep learning;
the coding and decoding structure based on deep learning comprises a first convolutional layer, a first cyclic residual block, a multi-level nested coding and decoding structure and a second convolutional layer which are connected in sequence; the multi-level nested coding and decoding structure is formed by nesting and inserting a next-level coding and decoding structure between a coder and a decoder of each level of coding and decoding structure;
in this embodiment, the first cyclic Residual block is implemented by a current Residual conv
As shown in fig. 2, a schematic diagram of a deep learning-based encoding and decoding structure in a catheter detection method for a vascular aneurysm interventional surgery navigation system according to the present invention, a multi-level nested encoding and decoding structure, an encoder and a decoder of which respectively include a plurality of encoding modules and a plurality of decoding modules; the coding module is connected with the corresponding peer decoding module through residual connection;
the layer 1 is a convolutional layer Conv. + ReLU, the layer 2 is a cyclic convolution Block Recurrent ResidualConv.Block, the layers 4, 6, 8 and 10 are cyclic convolution blocks Recurrent Residua Conv.Block in downsampling, the layers 12, 14, 16 and 18 are cyclic convolution blocks Recurrent Up-Conv.Block in upsampling, the layers 3, 5, 7 and 9 are coding residual blocks inversed Res-blocks, the layers 11, 13, 15 and 17 are decoding blocks Decode Block, the layer 19 is a convolutional layer Conv. + ReLU, and the layers A-H, B-G, C-F and D-E are coding and decoding structures corresponding to a mode of residual connection.
In this embodiment, the residual connection is realized through "short connection", so that the problems of gradient disappearance, overfitting and the like are solved, and the classification accuracy is greatly improved along with the deepening of the network.
The coding module is constructed based on a MobileNetV2 network and sequentially comprises a coding residual block and a second cyclic residual block;
as shown in fig. 3, in the schematic view of the structure of the encoded residual error block invoked Res-blocks in the catheter detection method for a vascular aneurysm interventional surgery navigation system of the present invention, in this embodiment, the encoded residual error block is implemented by replacing a feature extraction structure in a mobilene V2 structure with an invoked Res-blocks module that occupies fewer computing resources and has a faster computing speed; the module exists in a downsampling process and consists of a 1x1 convolution block, a 3x3 convolution block, a 1x1 convolution block and a Dwise convolution operation.
In the embodiment, the second cyclic Residual block is realized by a current Residual conj.Block, and the combination of the Inverted Res-blocks module and the current Residual conj.Block structure achieves better accuracy and real-time speed;
as shown in fig. 4, which is a schematic structural diagram of a cyclic Residual block recovery constant.block in the catheter detection method for a vascular aneurysm interventional surgery navigation system according to the present invention, the cyclic convolution is mainly performed by 2 convolution blocks of 3 × 3, so that context information can be better obtained.
As shown in fig. 5, in the catheter detection method for a vascular aneurysm interventional operation navigation system according to the present invention, the decoding Block Decoder Block is a decoding Block and a third cyclic residual Block in sequence based on a U-net neural structure and a recurrent neural network component.
In the embodiment, the decoding Block is realized by a Decoder Block to help recover the detail information of the image, and the decoding Block is formed by 1x1 convolution, 4x4 transposition convolution and 1x1 convolution;
in this embodiment, the third cyclic Residual Block is also implemented by a current Residual conj.
In this embodiment, as shown in fig. 6, the image of the catheter obtained by manual marking in the embodiment of the present invention, some central points are marked on the catheter, and a spline is fitted through these points to obtain the final target. For each image, the device was manually annotated by two technical experts with more than 5 years of medical imaging experience. A marker is considered valid when the average error distance of the centerlines of the catheters marked by both is less than 0.5 pixel. After annotation, each 2d x-ray image in the sequence is a binary image of size 512 x 512.
The manually annotated tokens are used in part as training and in part to verify the performance of the model.
Further, to address the limited number of catheter samples in this embodiment, a data enhancement algorithm (i.e., 0o < α <360o rotation, in the x and y directions) can be applied to our model.
Step S21, generating a first characteristic image for any picture to be detected in the video sequence to be detected through a first convolution layer in the coding and decoding structure based on the deep learning;
step S22, generating a second feature image from the first feature image through a first cyclic residual block in the coding and decoding structure based on the deep learning;
in this embodiment, the first cyclic Residual block is implemented by a current Residual conv.
Step S23, based on the second feature image, performing hierarchical coding through each coding module in the coding and decoding structure based on the deep learning to obtain a feature compressed image;
step S24, based on the feature compressed image, through each decoding module in the coding and decoding structure based on the deep learning, the hierarchical decoding is carried out, and the up-sampling feature image is obtained;
step S25, based on the up-sampling feature image, generating a binary segmentation mask of the catheter corresponding to the picture to be detected through the second convolution layer in the coding and decoding structure based on the deep learning, and obtaining a binary segmentation mask sequence of the catheter.
In this embodiment, the second cyclic residual block and the third cyclic residual block have the same structure and are residual blocks including two cyclic convolution layers.
In this embodiment, the encoded residual block is a residual block including a fourth convolutional layer, a Dwise convolutional layer, and a fifth convolutional layer.
In this embodiment, the decoding blocks are the sixth convolutional layer, the transposed convolutional layer, and the seventh convolutional layer.
In this embodiment, the sixth convolutional layer and the seventh convolutional layer may be convolutional layers in which batch regularization layers are incorporated.
In this embodiment, in the deep learning-based codec structure, the loss function used in the training is:
wherein L isBCEIs a binary cross entropy loss function, βiIs a coordination parameter of the gradient density,
is the pixel label of the ith pixel, 1 is the conduit, 0 is the background, piIs that
And finally, predicting the value.
In the embodiment, the coding and decoding structure based on deep learning is initialized by adopting a MSRA method; particularly a Gaussian distribution with the mean value of 0 and the variance of 2/n, and the adoption of the initialization mode can help the segmentation detection network to better converge.
In the embodiment, the loss function is reduced by a random gradient descent algorithm in the training of the deep learning-based coding and decoding structure, and the learning rate is reduced by 2 times when the verification precision reaches saturation after multiple iterations; the deep learning based codec structure of this embodiment is implemented using one NVIDIA TITAN Xp (12GB) in a PyTorch library (version 0.4.1). The initial learning rate was 0.001, the weight decay was 0.0005, and the momentum was 0.9. To find the best performance, the learning rate is reduced by a factor of 2 when the verification accuracy is saturated. In addition, the batch size was set to 32, and 300 epochs were used for each model training.
The present invention compares this with other segmentation methods and evaluates model performance in terms of sensitivity, F1 score and processing time. To calculate the processing time, the sequence is loaded into the proposed model and then each frame is calculated off-line in parallel. After obtaining the results (N) for all frames, the total time T may be calculated. Therefore, the final processing time per frame is 1000 × T/Nms.
Table 1 shows the effect comparison of the present invention with other segmentation methods;
to further verify model performance, the average of the distances from ground truth to the segmentation results was also evaluated and referred to as catheter distance error (CSP). And the like, and the more excellent results are obtained. Compared with other methods, the method has the problem of incomplete catheter segmentation, the method provided by the invention can accurately track the catheter in all group sequences, has no great problem of catheter loss, and has better robustness. The structure herein was found to make various improvements using ablation algorithms in the experiments.
Table 2 shows the performance comparison of the present invention with other segmentation methods;
the present invention proposes a new efficient network FRR-Net to address the challenging task of catheter segmentation and tracking for EVAR in 2D X fluoroscopy. Quantitative and qualitative assessment of the CTRSeg data set showed that the method of the present patent achieved significant improvements in both accuracy and robustness. As shown in fig. 6, a catheter image obtained by a manual marking method in the embodiment of the present invention, fig. 7, a catheter image obtained by using a U-net method in the prior art, and fig. 8, a catheter image obtained by using an RCL method in the prior art, the images of the catheter can be segmented by using the prior art, but the prior art has disadvantages, and as shown in fig. 9, a catheter image obtained by introducing a rapid encoding and decoding structure of a secure Residual is an embodiment of the present invention. Furthermore, the inference rate of our approach is approximately 16FPS, which is promising for real-time assisting physicians in EVAR.
Vandini et al propose a guide wire segmentation method with the best effect at present, and adopt a SEGlet robust feature improved curve fitting method to segment the guide wire, and SEGlets are introduced with features to overcome the problems of huge deformation in a continuous video sequence, but tracking loss, low speed and the like.
In terms of the detection and tracking of catheters, Ambrosini proposes an automated method that attempts to segment the entire catheter, but requires post-processing to join the segmented branches into a complete catheter, which is complicated and unnecessary.
In terms of detection tracking of catheters, Yatziv et al propose an improved fast marching method that combines clinical knowledge, limits the search space, first detects the catheter tip through a cascaded detector, and then tracks the catheter tip using an improved geodesic framework (using fast marching algorithms for weighted geodesic distances). The algorithm is mainly aimed at the interaction of other medical instruments and catheters in the operation, but needs manual initialization and is not suitable for the case of catheter tip occlusion.
Ma et al propose a fast speckle detection method, which mainly includes: fast blob detection algorithms, shape-based searching, and model-based detection. A catheter model is extracted according to a detection method, and the model is used as the input of a tracking method, so that various catheters can be detected simultaneously in real time. But the catheter tip is positioned facing the user while the algorithm assumes that the shape of the catheter is fixed, taking into account the fact that the catheter is deformed and that the C-arm is at an extreme angle.
Wu et al propose a catheter segmentation algorithm, which extracts an initial position of a catheter by using a Fast speckle detection algorithm in combination with a patch analysis method, then detects and tracks a catheter segment in a constrained search space based on a Fast-Up Robust Features (SURF) detector and a Fast-FD algorithm, and finally integrates and smoothes the catheter segment into a complete catheter by using a Kalman filter-based growth method and a hierarchical graph model. The method can automatically detect tracking, but the algorithm is only suitable for the case of small curvature of the catheter.
Ambrosini et al propose a catheter detection algorithm based on Hidden Markov Models (HMM) that obtains a 3-D vessel tree using 3-D rotational angiography images, and that tracks the catheter tip in the 3-D vessel tree to obtain its 3-D position using localization in the 2-D image based on the state transition probability distribution in the HMM Model. The algorithm obtains the 3-D position of the catheter primarily from the angiographic image, but there are some parameters to be determined in the algorithm.
Fazlali et al utilize multi-scale Top-Hat transformation to enhance each frame of image and calculate the blood vessel distribution map in each frame, utilize a guide filter, regard the catheter structure as a valley-shaped structure, utilize ridge detection algorithm and Hough transformation to detect the catheter in the first frame, then utilize second-order polynomial fitting catheter in the remaining image, the algorithm belongs to automatic detection and tracking algorithm, it is high to detect the accuracy.
Hoffmann et al use dual viewing angles to achieve 3-D tracking of guide wires, use a graph search algorithm to detect catheters in images of different viewing angles respectively, and then use detection results at two viewing angles to perform 3-D reconstruction on the catheters. The method can obtain the 3-D structure of the catheter, but the images under two visual angles are needed, only for the EP catheter, one seed point needs to be marked manually in each frame, and meanwhile, if the two catheters are overlapped, the detection result needs to be corrected manually. In the aspect of detection and tracking for the guide wire, based on a B-spline curve, Baert proposes an energy minimization guide wire tracking algorithm. But the algorithm requires forcing the smoothness of the curve and an additional penalty term to constrain the change in curve length.
Slabaugh et al propose a phase consistency algorithm, which comprises fitting a guide wire by using a B-spline curve model, evolving the motion of control points according to the phase consistency of the control points of a spline function to obtain new positions of the control points, and fitting the guide wire by using the control points of the new positions to achieve the purpose of tracking the guide wire, wherein the control points need to be reinitialized after the guide wire tracking fails.
Pauly et al obtains a motion distribution model of the guide wire by learning according to motion deformation of the guide wire by using a machine learning method, and the method has robustness on contrast change of images and partial shielding of complex backgrounds and interventional instruments, but does not limit motion regions of guide wires of adjacent frames in an algorithm.
Heibel et al propose algorithms based on Markov Random Field Models (MRF) that use discrete points as guide wire control points and then transform the guide wire tracking problem into a labeling problem that optimizes the curve made up of discrete points using a maximum posterior probability method. However, due to the low signal-to-noise ratio of the X-ray image, the method is easy to cause the lack of the guide wire, and the complexity is high when the guide wire is obtained according to the starting point.
Chang et al fit a guide wire using a B-spline model and a region-based probabilistic algorithm, where traditional B-spline curves rely primarily on control points for fitting, and the method relies primarily on the nodes of the curve.
Wang et al divides the guidewire into three sections, detects each section separately using a detector, and then combines the three sections using a bayesian network to form a complete guidewire. The method can solve the problem of any nonlinear motion of the guide wire, but needs to track by using manually marked training data, and is not suitable for being applied to C-shaped arms with different parameters.
The method for detecting the navigation instrument for the vascular aneurysm interventional operation is further deepened and improved on the basis of the patents, and obviously improves the detection speed, the detection precision, the robustness caused by individual difference and the like compared with the common instrument detection method.
The catheter detection system for a vascular aneurysm interventional surgery navigation system according to a second embodiment of the present invention comprises: a video acquisition unit 100, a mask generation unit 200, and a result generation unit 300;
a video acquisition unit 100, configured to acquire an X-ray transmission video sequence of a region including a catheter during a surgical procedure as a video sequence to be detected;
a mask generating unit 200, which generates a binary segmentation mask sequence of the catheter based on the video sequence to be detected through a trained coding and decoding structure based on deep learning;
and a result generating unit 300, which overlays the binary segmentation mask sequence on the video sequence to be detected to obtain a video sequence of the catheter.
In some preferred embodiments, the mask generating unit 200 includes:
a video extraction subunit 210, configured to generate, for any to-be-detected picture in the to-be-detected video sequence, a first feature image through a first convolution layer in the deep learning-based coding and decoding structure;
an image preprocessing subunit 220, configured to generate a second feature image from the first feature image through a first cyclic residual block in a deep learning-based codec structure;
an image encoding subunit 230, configured to perform hierarchical encoding through each encoding module in the deep learning-based encoding and decoding structure based on the second feature image, to obtain a feature compressed image;
an image decoding subunit 240, configured to perform hierarchical decoding on the feature compressed image through each decoding module in the deep learning-based coding and decoding structure, so as to obtain an upsampled feature image;
a mask generating subunit 250, configured to generate, based on the upsampled feature image, a binary segmentation mask of the catheter corresponding to the picture to be detected through the second convolutional layer in the deep learning-based coding and decoding structure, so as to obtain a binary segmentation mask sequence of the catheter.
It can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working process and related description of the system described above may refer to the corresponding process in the foregoing method embodiments, and will not be described herein again.
It should be noted that, the catheter detection system for a vascular aneurysm interventional operation navigation system provided in the above embodiment is only exemplified by the division of the above functional modules, and in practical applications, the above function distribution may be completed by different functional modules according to needs, that is, the modules or steps in the embodiment of the present invention are further decomposed or combined, for example, the modules in the above embodiment may be combined into one module, or may be further split into multiple sub-modules, so as to complete all or part of the above described functions. The names of the modules and steps involved in the embodiments of the present invention are only for distinguishing the modules or steps, and are not to be construed as unduly limiting the present invention.
A storage device according to a third embodiment of the present invention stores a plurality of programs, which are suitable for being loaded and executed by a processor to implement the above-mentioned catheter detection method for a vessel aneurysm interventional surgery navigation system.
A processing apparatus according to a fourth embodiment of the present invention includes a processor, a storage device; a processor adapted to execute various programs; a storage device adapted to store a plurality of programs; the program is suitable to be loaded and executed by a processor to implement the above-mentioned catheter detection method for a vascular aneurysm interventional surgery navigation system.
It can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes and related descriptions of the storage device and the processing device described above may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
Those of skill in the art would appreciate that the various illustrative modules, method steps, and modules described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that programs corresponding to the software modules, method steps may be located in Random Access Memory (RAM), memory, Read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. To clearly illustrate this interchangeability of electronic hardware and software, various illustrative components and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as electronic hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
The terms "first," "second," and the like are used for distinguishing between similar elements and not necessarily for describing or implying a particular order or sequence.
The terms "comprises," "comprising," or any other similar term are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
So far, the technical solutions of the present invention have been described in connection with the preferred embodiments shown in the drawings, but it is easily understood by those skilled in the art that the scope of the present invention is obviously not limited to these specific embodiments. Equivalent changes or substitutions of related technical features can be made by those skilled in the art without departing from the principle of the invention, and the technical scheme after the changes or substitutions can fall into the protection scope of the invention.
Claims (13)
1. A catheter detection method for a vascular aneurysm interventional surgery navigation system, the detection method comprising:
step S10, acquiring an X-ray transmission video sequence of an area containing a catheter in the operation process as a video sequence to be detected;
step S20, based on the video sequence to be detected, generating a binary segmentation mask sequence of the catheter through a trained coding and decoding structure based on deep learning;
step S30, covering the binary segmentation mask sequence on the video sequence to be detected to obtain a video sequence of the catheter;
the coding and decoding structure based on deep learning comprises a first convolutional layer, a first cyclic residual block, a multi-level nested coding and decoding structure and a second convolutional layer which are connected in sequence; the multi-level nested coding and decoding structure is formed by nesting and inserting a next-level coding and decoding structure between a coder and a decoder of each level of coding and decoding structure;
the encoder and the decoder of the multi-level nested coding and decoding structure respectively comprise a plurality of coding modules and a plurality of decoding modules; the coding module is connected with the corresponding peer decoding module through residual connection;
the encoding module is constructed based on a MobileNet V2 network and comprises an encoding residual block and a second cyclic residual block which are connected in sequence;
the decoding module is constructed based on a U-net neural structure and a recurrent neural network and comprises a decoding block and a second cyclic residual block which are connected in sequence.
2. The catheter detection method for a vascular aneurysm interventional procedure navigation system according to claim 1, wherein step S20 includes:
step S21, generating a first characteristic image for any picture to be detected in the video sequence to be detected through a first convolution layer in the coding and decoding structure based on the deep learning;
step S22, generating a second feature image from the first feature image through a first cyclic residual block in the coding and decoding structure based on the deep learning;
step S23, based on the second feature image, performing hierarchical coding through each coding module in the coding and decoding structure based on the deep learning to obtain a feature compressed image;
step S24, based on the feature compressed image, through each decoding module in the coding and decoding structure based on the deep learning, the hierarchical decoding is carried out, and the up-sampling feature image is obtained;
step S25, based on the up-sampling feature image, generating a binary segmentation mask of the catheter corresponding to the picture to be detected through the second convolution layer in the coding and decoding structure based on the deep learning, and obtaining a binary segmentation mask sequence of the catheter.
3. The catheter detection method for a vascular aneurysm interventional procedure navigation system of claim 1, wherein the second cyclic residual block is a residual block comprising two cyclic convolution layers.
4. The catheter detection method for a vascular aneurysm interventional procedure navigation system according to claim 1, wherein the encoded residual block is a residual block including a fourth convolutional layer, a Dwise convolutional layer, and a fifth convolutional layer.
5. The catheter detection method for a vascular aneurysm interventional procedure navigation system of claim 1, wherein the decoding block comprises a sixth convolutional layer, a transposed convolutional layer, and a seventh convolutional layer, which are connected in sequence.
6. The catheter detection method for a vascular aneurysm interventional procedure navigation system of claim 5, wherein the sixth and seventh convolutional layers are convolutional layers incorporating a batch of regularization layers.
7. The catheter detection method for a vascular aneurysm interventional procedure navigation system according to claim 1, wherein the deep learning based codec structure is trained with a loss function of:
wherein L isBCEIs a binary cross entropy loss function, βiIs gradient density coordinationThe parameters are set to be in a predetermined range,
is the pixel label of the ith pixel, 1 is the conduit, 0 is the background, piIs that
And finally, predicting the value.
8. The catheter detection method for a vascular aneurysm interventional surgery navigation system according to claim 1, wherein the deep learning based codec structure is initialized by a method of MSRA.
9. The method as claimed in claim 1, wherein the deep learning based codec structure reduces the learning rate by 2 times when the verification accuracy is saturated after multiple iterations in the training.
10. A catheter detection system for a vascular aneurysm interventional procedure navigation system, the detection system comprising: a video acquisition unit 100, a mask generation unit 200, and a result generation unit 300;
a video acquisition unit 100, configured to acquire an X-ray transmission video sequence of a region including a catheter during a surgical procedure as a video sequence to be detected;
a mask generating unit 200, which generates a binary segmentation mask sequence of the catheter based on the video sequence to be detected through a trained coding and decoding structure based on deep learning;
and a result generating unit 300, which overlays the binary segmentation mask sequence on the video sequence to be detected to obtain a video sequence of the catheter.
11. The catheter detection system for a vascular aneurysm interventional procedure navigation system according to claim 10, wherein the mask generation unit 200 comprises:
a video extraction subunit 210, configured to generate, for any to-be-detected picture in the to-be-detected video sequence, a first feature image through a first convolution layer in the deep learning-based coding and decoding structure;
an image preprocessing subunit 220, configured to generate a second feature image from the first feature image through a first cyclic residual block in a deep learning-based codec structure;
an image encoding subunit 230, configured to perform hierarchical encoding through each encoding module in the deep learning-based encoding and decoding structure based on the second feature image, to obtain a feature compressed image;
an image decoding subunit 240, configured to perform hierarchical decoding on the feature compressed image through each decoding module in the deep learning-based coding and decoding structure, so as to obtain an upsampled feature image;
a mask generating subunit 250, configured to generate, based on the upsampled feature image, a binary segmentation mask of the catheter corresponding to the picture to be detected through the second convolutional layer in the deep learning-based coding and decoding structure, so as to obtain a binary segmentation mask sequence of the catheter.
12. A storage device having stored thereon a plurality of programs, wherein the programs are adapted to be loaded and executed by a processor to implement the catheter detection method for a vessel aneurysm interventional procedure navigation system according to claims 1-9.
13. A processing apparatus comprising a processor adapted to execute programs; and a storage device adapted to store a plurality of programs; wherein the program is adapted to be loaded and executed by a processor to perform: the catheter detection method for a vascular aneurysm interventional procedure navigation system of claims 1-9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010546190.2A CN111681254A (en) | 2020-06-16 | 2020-06-16 | Catheter detection method and system for vascular aneurysm interventional operation navigation system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010546190.2A CN111681254A (en) | 2020-06-16 | 2020-06-16 | Catheter detection method and system for vascular aneurysm interventional operation navigation system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111681254A true CN111681254A (en) | 2020-09-18 |
Family
ID=72436239
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010546190.2A Pending CN111681254A (en) | 2020-06-16 | 2020-06-16 | Catheter detection method and system for vascular aneurysm interventional operation navigation system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111681254A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112116608A (en) * | 2020-10-22 | 2020-12-22 | 上海联影医疗科技股份有限公司 | Guide wire segmentation method and device, electronic equipment and storage medium |
| CN112348883A (en) * | 2020-11-24 | 2021-02-09 | 中国科学院自动化研究所 | Interventional instrument endpoint real-time positioning system, method and device in vascular interventional operation |
| CN112837218A (en) * | 2021-01-14 | 2021-05-25 | 中国科学院自动化研究所 | Aneurysm image analysis system, method and equipment based on double-flow boundary perception |
| CN113066061A (en) * | 2021-03-24 | 2021-07-02 | 同心医联科技（北京）有限公司 | Aneurysm detection method, system, terminal and medium based on MRA |
| CN114862865A (en) * | 2022-07-11 | 2022-08-05 | 天津大学 | Vessel segmentation method and system based on multi-view coronary angiography sequence image |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109544662A (en) * | 2018-11-29 | 2019-03-29 | 文创智慧科技（武汉）有限公司 | A kind of animation style line original text painting methods and system based on SRUnet |
| CN109886273A (en) * | 2019-02-26 | 2019-06-14 | 四川大学华西医院 | A kind of CMR classification of image segmentation system |
| CN110706232A (en) * | 2019-09-29 | 2020-01-17 | 五邑大学 | Texture image segmentation method, electronic device and computer storage medium |
| CN111127490A (en) * | 2019-12-31 | 2020-05-08 | 杭州电子科技大学 | Medical image segmentation method based on cyclic residual U-Net network |
| CN111145186A (en) * | 2019-12-17 | 2020-05-12 | 中国科学院深圳先进技术研究院 | Neural network structure, image segmentation method, device and storage medium |
-
2020
- 2020-06-16 CN CN202010546190.2A patent/CN111681254A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109544662A (en) * | 2018-11-29 | 2019-03-29 | 文创智慧科技（武汉）有限公司 | A kind of animation style line original text painting methods and system based on SRUnet |
| CN109886273A (en) * | 2019-02-26 | 2019-06-14 | 四川大学华西医院 | A kind of CMR classification of image segmentation system |
| CN110706232A (en) * | 2019-09-29 | 2020-01-17 | 五邑大学 | Texture image segmentation method, electronic device and computer storage medium |
| CN111145186A (en) * | 2019-12-17 | 2020-05-12 | 中国科学院深圳先进技术研究院 | Neural network structure, image segmentation method, device and storage medium |
| CN111127490A (en) * | 2019-12-31 | 2020-05-08 | 杭州电子科技大学 | Medical image segmentation method based on cyclic residual U-Net network |
Non-Patent Citations (5)
| Title |
|---|
| HAILAN CHENG等: "Modified U-Net block network for lung nodule detection", 《2019 IEEE 8TH JOINT INTERNATIONAL INFORMATION TECHNOLOGY AND ARTIFICIAL INTELLIGENCE CONFERENCE (ITAIC 2019)》 * |
| MARK SANDLER等: "MobileNetV2: Inverted Residuals and Linear Bottlenecks", 《2018 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION》 * |
| VLADIMIR I. IGLOVIKOV等: "TernausNetV2: Fully Convolutional Network for Instance Segmentation", 《ARXIV:1806.00844V2 [CS.CV] 19 JUN 2018》 * |
| YAN-JIE ZHOU等: "FRR-NET: Fast Recurrent Residual Networks for Real-Time Catheter Segmentation and Tracking in Endovascular Aneurysm Repair", 《《2020 IEEE 17TH INTERNATIONAL SYMPOSIUM ON BIOMEDICAL IMAGING (ISBI)》》 * |
| 李锵等: "MＲI 脑肿瘤图像分割研究进展及挑战", 《中国图象图形学报》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112116608A (en) * | 2020-10-22 | 2020-12-22 | 上海联影医疗科技股份有限公司 | Guide wire segmentation method and device, electronic equipment and storage medium |
| CN112116608B (en) * | 2020-10-22 | 2022-10-14 | 上海联影医疗科技股份有限公司 | Guide wire segmentation method and device, electronic equipment and storage medium |
| CN112348883A (en) * | 2020-11-24 | 2021-02-09 | 中国科学院自动化研究所 | Interventional instrument endpoint real-time positioning system, method and device in vascular interventional operation |
| CN112348883B (en) * | 2020-11-24 | 2021-06-29 | 中国科学院自动化研究所 | Interventional instrument endpoint real-time positioning system, method and device in vascular interventional operation |
| CN112837218A (en) * | 2021-01-14 | 2021-05-25 | 中国科学院自动化研究所 | Aneurysm image analysis system, method and equipment based on double-flow boundary perception |
| CN113066061A (en) * | 2021-03-24 | 2021-07-02 | 同心医联科技（北京）有限公司 | Aneurysm detection method, system, terminal and medium based on MRA |
| CN113066061B (en) * | 2021-03-24 | 2024-02-02 | 同心医联科技（北京）有限公司 | MRA-based aneurysm detection method, system, terminal and medium |
| CN114862865A (en) * | 2022-07-11 | 2022-08-05 | 天津大学 | Vessel segmentation method and system based on multi-view coronary angiography sequence image |
| CN114862865B (en) * | 2022-07-11 | 2022-09-06 | 天津大学 | Vessel segmentation method and system based on multi-view coronary angiography sequence image |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111681254A (en) | Catheter detection method and system for vascular aneurysm interventional operation navigation system | |
| CN108633312B (en) | Contrast cloud detection method in X-ray image | |
| Wu et al. | Automatic detection of coronary artery stenosis by convolutional neural network with temporal constraint | |
| US8126241B2 (en) | Method and apparatus for positioning a device in a tubular organ | |
| US8588501B2 (en) | Automatic pose initialization for accurate 2-D/3-D registration applied to abdominal aortic aneurysm endovascular repair | |
| US8620050B2 (en) | System and method for 2-D/3-D registration between 3-D volume and 2-D angiography | |
| Hao et al. | Sequential vessel segmentation via deep channel attention network | |
| CN112997215A (en) | Method and system for automatically regionalizing blood vessels of medical images by using machine learning and image processing algorithms | |
| US9058664B2 (en) | 2D-2D fusion for interventional guidance in trans-catheter aortic valve implantation | |
| CN111080700A (en) | Medical instrument image detection method and device | |
| JP2018536235A (en) | Shadow detection and verification in intravascular images | |
| Rezaei et al. | Whole heart and great vessel segmentation with context-aware of generative adversarial networks | |
| CN112348821B (en) | Guide wire segmentation and tip point positioning method, system and device based on X-ray image | |
| Wu et al. | Fast catheter segmentation from echocardiographic sequences based on segmentation from corresponding X-ray fluoroscopy for cardiac catheterization interventions | |
| JP2022539078A (en) | Systems and methods for guidance of motion conditioning devices using vascular roadmaps | |
| US9186087B2 (en) | Combined cardiac and respiratory motion compensation for atrial fibrillation ablation procedures | |
| CN112348883B (en) | Interventional instrument endpoint real-time positioning system, method and device in vascular interventional operation | |
| CN111724365B (en) | Interventional instrument detection method, system and device for endovascular aneurysm repair operation | |
| CN116503607B (en) | CT image segmentation method and system based on deep learning | |
| CN105894445B (en) | A kind of coronary artery image processing method and device | |
| CN111985485A (en) | Pyramid attention cycle network-based surgical interventional instrument tracking method | |
| CN114511581A (en) | Multi-task multi-resolution collaborative esophageal cancer lesion segmentation method and device | |
| Blondel et al. | Automatic trinocular 3D reconstruction of coronary artery centerlines from rotational X-ray angiography | |
| JP5364009B2 (en) | Image generating apparatus, image generating method, and program thereof | |
| Brost et al. | Catheter tracking: Filter-based vs. learning-based |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200918 |