CN110517243A - A kind of localization method and system based on DSA image - Google Patents
A kind of localization method and system based on DSA image Download PDFInfo
- Publication number
- CN110517243A CN110517243A CN201910782514.XA CN201910782514A CN110517243A CN 110517243 A CN110517243 A CN 110517243A CN 201910782514 A CN201910782514 A CN 201910782514A CN 110517243 A CN110517243 A CN 110517243A
- Authority
- CN
- China
- Prior art keywords
- target area
- image
- model
- region
- dimentional
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30016—Brain
-
- 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
- G06T2207/30096—Tumor; Lesion
-
- 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
This specification embodiment discloses a kind of localization method and system based on DSA image, by the positioning to target area in two-dimentional DSA sequence image to be processed, solve the problems, such as that " observation method of naked eye " is influenced by the subjective consciousness bigger, cost more time.The localization method includes: from two-dimentional DSA sequence image to be processed, to obtain the frame containing target area using the first model;Extract the doubtful localization region of target area;Using third model, the image in suspected target region is obtained;Using the 4th model, the location information of target area is obtained;The location information of target area is reverted to the frame where target area, obtains localization region of the target area in two-dimentional DSA sequence image to be processed.The localization method and system based on DSA image that this specification embodiment provides can be realized the target area directly displayed in two-dimentional DSA sequence image, reduce the time of artificial observation, thinking and judgement, improve accuracy of judgement degree.
Description
Technical field
This specification is related to medical image and field of computer technology more particularly to a kind of positioning side based on DSA image
Method and system.
Background technique
Intracranial aneurysm is a kind of common vascular conditions, which is since the local anomaly of entocranial artery inner cavity expands
A kind of strumae of arterial wall caused by.It is reported that illness rate of the encephalic Unruptured aneurysm in China adult is high
Up to 7%, after encephalic Unruptured aneurysm ruptures, it is even dead to will lead to handicap.Therefore, entocranial artery is found early
Tumor is of great significance.
DSA (Digital subtraction angiography, digital subtraction angiography) is used as entocranial artery blood vessel
The goldstandard of deformity and Diagnosis of Aneurysm, is widely applied in clinic.It is seen currently, the positioning of intracranial aneurysm relies primarily on naked eyes
It examines and is judged." observation method of naked eye " is somebody's turn to do by reading two dimension DSA sequence image, tentatively judges whether there is intracranial aneurysm.
This method is influenced bigger by the observation visual angle of two-dimentional DSA sequence image and the subjective consciousness of observer, is easy to appear and is failed to pinpoint a disease in diagnosis,
And during observing, the thinking of observer is needed, the more time is spent.
Therefore, it is necessary to a kind of new localization methods, can exclude or reduce subjective factor and image documentation equipment imaging difference band
The diagnosis difference come, reduces the time of artificial observation, thinking and judgement, and can be improved the accuracy of judgement, as calculating
Machine householder method carries out diagnosis for later use DSA image and teaching research provides foundation.
Summary of the invention
This specification embodiment provides a kind of localization method and system based on DSA image, asks for solving following technology
Topic: needing a kind of new localization method, can exclude or reduce subjective factor and the diagnosis of image documentation equipment imaging difference bring is poor
It is different, reduce the time of artificial observation, thinking and judgement, and can be improved the accuracy of judgement, as computer-aid method,
Diagnosis is carried out for later use DSA image and teaching research provides foundation.
This specification embodiment provides a kind of localization method based on DSA image, comprising the following steps:
Using the first model, from two-dimentional DSA sequence image to be processed, the frame containing target area is obtained, wherein institute
Stating the first model is the model being obtained ahead of time based on deep learning method, and the two-dimentional DSA sequence image to be processed is multiframe
's;
Based on the frame containing target area, the doubtful localization region of target area is extracted, wherein the doubtful positioning
Region is made of a series of doubtful zonules containing target area;
Using third model, classify to the image of the doubtful localization region, obtain the image in suspected target region,
Wherein, the third model is two disaggregated models being obtained ahead of time based on deep learning method, and the suspected target region is base
In the possible target area of third model output;
Using the 4th model, the image containing the suspected target region is split, the positioning of target area is obtained
Information, wherein the 4th model is the model being obtained ahead of time based on deep learning method;
The location information of the target area is reverted to the frame where target area, obtains target area described wait locate
Localization region in the two-dimentional DSA sequence image of reason.
Further, the method also includes pre-treatment steps:
The two-dimentional DSA sequence image to be processed is normalized, image transformation, scaling, interception and filling
One of processing or a variety of processing, make the picture size of the two-dimentional DSA sequence image to be processed and first model
And pel spacing is consistent.
Further, described to utilize the first model, from two-dimentional DSA sequence image to be processed, acquisition contains target area
The frame in domain, specifically includes:
The two-dimentional DSA sequence image to be processed is inputted into the first model, obtains and contains in the two dimension DSA sequence image
There is the frame of target area.
Further, described based on the frame containing target area, the doubtful localization region of target area is extracted, specifically
Include:
Using the second model, from the frame containing target area, determine that the frame containing target area is corresponding
Temperature figure;
According to the second preset threshold, each pixel for belonging to target area in the temperature figure is determined;
Based on the coordinate for each pixel for belonging to target area in the temperature figure, from the frame containing target area
The corresponding region of the coordinate is intercepted, the doubtful localization region of target area is obtained.
Further, described to utilize third model, classify to the image of the doubtful localization region, obtains doubtful mesh
The image for marking region, specifically includes:
The third model according to the characteristics of image of the doubtful localization region, judge the doubtful localization region whether be
Target area exports the image in suspected target region.
Further, described to utilize the 4th model, the image containing the suspected target region is split, mesh is obtained
The location information for marking region, specifically includes:
The image in the suspected target region is inputted the 4th model to be split, the 4th model output contains
The image of target area and corresponding coordinate information, obtain the location information of target area.
Further, the location information by the target area reverts to the frame where target area, obtains target
Localization region of the region in the two-dimentional DSA sequence image to be processed, specifically includes:
The coordinate points of location information based on the target area, by linear transformation, by the positioning of the target area
Information reverts to the frame where the target area, obtains the location information of the target area where the target area
The corresponding coordinate points of frame;
Corresponding coordinate points by the location information of the target area in the frame where the target area carry out interpolation
Line obtains localization region of the target area in the two-dimentional DSA sequence image to be processed.
A kind of positioning system based on DSA image that this specification embodiment provides, comprising:
Receiving unit receives two-dimentional DSA sequence image to be processed;
Processing unit positions the two-dimentional DSA sequence image to be processed;
Output unit shows the positioning result of the two-dimentional DSA sequence image to be processed.
Further, described that the two-dimentional DSA sequence image to be processed is positioned, it specifically includes:
Using the first model, from two-dimentional DSA sequence image to be processed, the frame containing target area is obtained, wherein institute
Stating the first model is the model being obtained ahead of time based on deep learning method, and the two-dimentional DSA sequence image to be processed is multiframe
's;
Based on the frame containing target area, the doubtful localization region of target area is extracted, wherein the doubtful positioning
Region is made of a series of doubtful zonules containing target area;
Using third model, classify to the image of the doubtful localization region, obtain the image in suspected target region,
Wherein, the third model is two disaggregated models being obtained ahead of time based on deep learning method, and the suspected target region is base
In the possible target area of third model output;
Using the 4th model, the image containing the suspected target region is split, the positioning of target area is obtained
Information, wherein the 4th model is the model being obtained ahead of time based on deep learning method;
The location information of the target area is reverted to the frame where target area, obtains target area described wait locate
Localization region in the two-dimentional DSA sequence image of reason.
Further, the method also includes pre-treatment steps:
The two-dimentional DSA sequence image to be processed is normalized, image transformation, scaling, interception and filling
One of processing or a variety of processing, make the picture size of the two-dimentional DSA sequence image to be processed and first model
And pel spacing is consistent.
Further, described to utilize the first model, from two-dimentional DSA sequence image to be processed, acquisition contains target area
The frame in domain, specifically includes:
The two-dimentional DSA sequence image to be processed is inputted into the first model, obtains and contains in the two dimension DSA sequence image
There is the frame of target area.
Further, described based on the frame containing target area, the doubtful localization region of target area is extracted, specifically
Include:
Using the second model, from the frame containing target area, determine that the frame containing target area is corresponding
Temperature figure;
According to the second preset threshold, each pixel for belonging to target area in the temperature figure is determined;
Based on the coordinate for each pixel for belonging to target area in the temperature figure, from the frame containing target area
The corresponding region of the coordinate is intercepted, the doubtful localization region of target area is obtained.
Further, described to utilize third model, classify to the image of the doubtful localization region, obtains doubtful mesh
The image for marking region, specifically includes:
The third model according to the characteristics of image of the doubtful localization region, judge the doubtful localization region whether be
Target area exports the image in suspected target region.
Further, described to utilize the 4th model, the image containing the suspected target region is split, mesh is obtained
The location information for marking region, specifically includes:
The image in the suspected target region is inputted the 4th model to be split, the 4th model output contains
The image of target area and corresponding coordinate information, obtain the location information of target area.
Further, the location information by the target area reverts to the frame where target area, obtains target
Localization region of the region in the two-dimentional DSA sequence image to be processed, specifically includes:
The coordinate points of location information based on the target area, by linear transformation, by the positioning of the target area
Information reverts to the frame where the target area, obtains the location information of the target area where the target area
The corresponding coordinate points of frame;
Corresponding coordinate points by the location information of the target area in the frame where the target area carry out interpolation
Line obtains localization region of the target area in the two-dimentional DSA sequence image to be processed.
This specification embodiment use at least one above-mentioned technical solution can reach it is following the utility model has the advantages that
This specification embodiment is by obtaining the frame containing target area, mentioning from two-dimentional DSA sequence image to be processed
The doubtful localization region for taking target area obtains suspected target region and is split, to obtain target area by classification
Location information, the location information of target area is passed through into linear transformation, to realize target area in two dimension DSA sequence image
Positioning, the present invention can be realized the target area directly displayed in two-dimentional DSA sequence image, exclude or reduce subjective factor and shadow
As equipment imaging difference bring diagnosis difference, reduces artificial observation, thinking and the time judged, improve accuracy of judgement degree.
Detailed description of the invention
In order to illustrate more clearly of this specification embodiment or technical solution in the prior art, below will to embodiment or
Attached drawing needed to be used in the description of the prior art is briefly described, it should be apparent that, the accompanying drawings in the following description is only
The some embodiments recorded in this specification, for those of ordinary skill in the art, in not making the creative labor property
Under the premise of, it is also possible to obtain other drawings based on these drawings.
Fig. 1 is a kind of schematic diagram for localization method based on DSA image that this specification embodiment provides;
Fig. 2 is the flow chart for frame of the acquisition containing target area that this specification embodiment provides;
Fig. 3 is the flow chart of the doubtful localization region for the extraction target area that this specification embodiment provides;
Fig. 4 is the flow chart in the acquisition suspected target region that this specification provides;
Fig. 5 is the flow chart being split to the image in suspected target region that this specification embodiment provides;
Fig. 6 is the positioning result schematic diagram based on two-dimentional DSA sequence image that this specification embodiment provides;
Fig. 7 is a kind of positioning system based on DSA image that this specification embodiment provides.
Specific embodiment
In order to make those skilled in the art more fully understand the technical solution in this specification, below in conjunction with this explanation
Attached drawing in book embodiment is clearly and completely described the technical solution in this specification embodiment, it is clear that described
Embodiment be merely a part but not all of the embodiments of the present application.Based on this specification embodiment, this field
Those of ordinary skill's every other embodiment obtained without creative efforts, all should belong to the application
The range of protection.
Fig. 1 is the schematic diagram of a kind of localization method based on DSA image that this specification embodiment provides, specifically include with
Lower step:
Step S101: utilizing the first model, and from two-dimentional DSA sequence image to be processed, acquisition contains target area
Frame.
DSA (Digital subtraction angiography, digital subtraction angiography) is to inject contrast agent to need
In the blood vessel to be checked, make exposed vessel original shape.DSA image is mainly used for observing the positioning measurement of vascular lesion, hemadostewnosis
And image is provided for interventional therapy.In practical applications, the two-dimentional DSA sequence image of observed person is multiframe, general to wrap
The image for including cranium brain normotopia and cranium brain side position, may also include cranium brain loxosis image.From the two-dimentional DSA sequence image of observed person
Determine the positioning of target area, it is necessary first to determine the frame containing target area.It, will be in one embodiment of this specification
The two-dimentional DSA sequence image of processing inputs the first model, and the first model, which exports in two-dimentional DSA sequence image to be processed, contains mesh
Mark the frame in region.In the another embodiment of this specification, two-dimentional DSA sequence image to be processed is inputted into the first mould
Type, the first model export the frame containing target area and two-dimentional DSA sequence to be processed in two-dimentional DSA sequence image to be processed
Image contains the probability of target area.It should be strongly noted that may have mesh in the multiframe of two-dimentional DSA sequence image
Region is marked, therefore the frame containing target area exported is a frame or multiframe, if there is no target area in two dimension DSA sequence image
Domain does not export the frame containing target area then.In addition, in practical applications, it can be according to the two-dimentional DSA sequence to be processed of output
The probability that column image contains target area provides reference to observer.Specifically, according to the first preset threshold, to two-dimentional DSA sequence
Probability further progress judgement in column image containing target area, to determine whether two dimension DSA sequence image contains target area
Domain.It is generally believed that two-dimentional DSA sequence image contains target when the aforementioned probability containing target area is higher than the first preset threshold
Region.In the specific implementation process, the first preset threshold can be set to 60%, it is believed that the probability value containing target area is higher than
When 60%, two-dimentional DSA sequence image contains target area.
In the embodiment of this specification, target area be can be according to scene and/or preset need is preset, and be preassigned
Interested region.In practical applications, target area can include but is not limited to: intracranial aneurysm, arteriovenous malformation.
The first model in this specification embodiment is the model for first passing through the training of deep learning method in advance and obtaining, for more
It is readily appreciated that and obtains the frame containing target area using the first model, be described in detail below and obtain the frame containing target area,
It is specific as shown in Figure 2.Fig. 2 is the flow chart for frame of the acquisition containing target area that this specification embodiment provides, and is specifically included:
Step S201: input two dimension DSA sequence image to convolutional neural networks.
Sample for carrying out the first model training is the two-dimentional DSA sequence image containing target area, and to two-dimentional DSA
Sequence image is labeled.For the accuracy for guaranteeing the first model of training, the quantity of training sample should be sufficiently large.
Step S203: first model of the training based on convolutional neural networks.
Abovementioned steps S201 two-dimentional DSA sequence image input convolutional neural networks after, to convolutional neural networks model into
Row training, optimization, obtain the first model.It may be implemented using first model: two-dimentional DSA sequence image inputted into the first model
Afterwards, the probability containing target area in two dimension DSA sequence image is exported.Optionally, two-dimentional DSA sequence image is inputted into the first mould
After type, output two dimension DSA sequence image contains the probability of target area and the frame containing target area.
Step S205: two-dimentional DSA sequence image to be processed is pre-processed.
In the specific implementation process, the frame number of two-dimentional DSA sequence image differs, usually between 10~40 frames, DSA image
File format be DICOM format.Since two-dimentional DSA sequence image is there may be differences such as sizes, by two-dimentional DSA sequence
It before column image inputs the first model, needs that two-dimentional DSA sequence image to be processed is normalized, image transformation, put
One of contracting, interception and filling processing or a variety of processing.Specifically, size is carried out to two-dimentional DSA sequence image to be processed
Scaling and/or pixel value normalization and/or pel spacing normalization and/or image transformation and/or scaling and/or interception and/or
The processing such as filling processing, make each frame image of pending two dimension DSA keep identical size and pel spacing, and with
Size and the pel spacing for carrying out the image of the sample of the first model training are consistent, so that two-dimentional DSA sequence image normally inputs
In first model.
Step S207: the two-dimentional DSA sequence image of pretreatment processing is inputted into the first model, obtains two dimension DSA sequence chart
The frame containing target area as in.
The first model of abovementioned steps S203 can determine the frame containing target area in two-dimentional DSA sequence image.Will before
Step S205 is stated after pretreated two dimension DSA sequence image inputs the first model, exports and contains in two dimension DSA sequence image
The frame of target area.In one embodiment of this specification, eight two dimension DSA sequence images containing four blood vessels, including
The two-dimentional DSA sequence image of normotopia and side position after two-dimentional DSA sequence image is normalized, inputs the first model, output
Frame containing target area in four blood vessel sequence images, that is, the frame containing target area in two dimension DSA sequence image.At this
In the another embodiment of specification, eight two dimension DSA sequence images containing four blood vessels, two including normotopia and side position
DSA sequence image is tieed up, after two-dimentional DSA sequence image is normalized, the first model is inputted, exports the two dimension of four blood vessels
Probability in DSA sequence image containing target area and the frame containing target area, that is, contain mesh in two dimension DSA sequence image
Mark the probability in region and the frame containing target area.
Step S103: based on the frame containing target area, the doubtful localization region of target area is extracted.
Frame containing target area is obtained using abovementioned steps S101, is needed further exist for from the frame containing target area
In, extract the doubtful localization region of target area.Fig. 3 is the doubtful positioning for the extraction target area that this specification embodiment provides
The flow chart in region, specifically includes:
Step S301: two dimension DSA sequence image and data label are inputted to convolutional neural networks.
Sample for carrying out the second model training is the two-dimentional DSA sequence image containing target area, and according to observation
The habit of person, label target region, other regions not marked are as background, using target area and background as data label,
Input convolutional neural networks are trained, which can be but be not limited to: the convolutional Neural net of 100 layer depths
Network.It should be strongly noted that the accuracy in order to guarantee training pattern, is sufficiently large for trained sample size.
Step S303: second model of the training based on convolutional neural networks.
Characteristics of image in the two-dimentional DSA sequence image and data label of extraction step S301 input, had been embodied
Cheng Zhong, characteristics of image can be but not limited to: the grayscale information of image, the grayscale information of neighbor pixel, gradient information, adjacent
The gradient information of pixel.Using the above method, convolutional neural networks model is trained, is optimized, obtains the second model.Benefit
It may be implemented with second model: after two-dimentional DSA sequence image inputs the second model, two-dimentional DSA sequence image pair can be exported
The temperature figure answered, and contain the probability that each pixel contains target area in two dimension DSA sequence image.
Step S305: utilizing the second model, determines the corresponding temperature figure of the frame containing target area.
The second model that frame input above mentioned step S3 03 in two-dimentional DSA sequence image containing target area are obtained, the
Two models export the corresponding temperature figure of the frame containing target area, and belong to the probability of target area containing each pixel.Specifically
The two-dimentional DSA sequence image of 1024*1024 can be exported the temperature figure of several 26*26, and each of temperature figure by ground
Pixel has a corresponding probability value.After the two-dimentional DSA sequence image of 1024*1024 inputs the second model, output is
The zonule of the corresponding 224*224 of two-dimentional DSA sequence image of 1024*1024, each pixel is corresponding general in temperature figure
Rate value indicates that the zonule of the corresponding 224*224 of two-dimentional DSA sequence image of 1024*1024 belongs to the probability of target area, small
Region indicates a part of region in two dimension DSA sequence image, and two-dimentional DSA sequence image is made of several zonules.
Step S307: according to the second preset threshold, each pixel for belonging to target area in temperature figure is determined.
The probability value of each pixel in the corresponding temperature figure of the frame containing target area obtained based on above mentioned step S3 05,
When the probability value of pixel is more than or equal to the second preset threshold, which is the pixel for belonging to target area, works as pixel
When the probability value of point is less than the second preset threshold, which is not belonging to the pixel of target area.In the specific implementation process,
Second preset threshold can be 60%.The method provided using this specification, can filter out in temperature figure and belong to target area
Pixel, which represents the pixel for belonging to target area in two dimension DSA sequence image.
Step S309: the coordinate based on each pixel for belonging to target area in temperature figure extracts the doubtful of target area
Localization region.
It is zonule since the pixel in temperature figure is corresponding, in the specific implementation process, needs further really
Determine position of the zonule in two-dimentional DSA sequence image.Specifically, belong to mesh in the temperature figure obtained based on above mentioned step S3 07
The coordinate for marking each pixel in region, interception belongs to the coordinate pair of each pixel of target area from the frame containing target area
The region answered comes out so that zonule be intercepted from the frame containing target area, obtains the doubtful localization region of target area.
Step S105: utilizing third model, classify to doubtful localization region, obtains suspected target region.
For the accuracy for guaranteeing positioning result, needs to obtain doubtful localization region further progress to abovementioned steps S103 and sentence
It is disconnected, determine suspected target region.Fig. 4 is the flow chart in the acquisition suspected target region that this specification provides, and is specifically included:
Step S401: input training sample to convolutional neural networks.
Sample for carrying out third model training is largely containing the cell containing target area of preset data label
Training sample input convolutional neural networks are trained by the image in domain, which can be but be not limited to: 100
The convolutional neural networks of layer depth.In the specific implementation process, preset data label can indicate that target is not present in background with 0
Region, 1 indicates that there are target areas.It should be strongly noted that the accuracy in order to guarantee training pattern, for trained sample
This quantity is sufficiently large.
Step S403: third model of the training based on convolutional neural networks.
The characteristics of image of the training sample of extraction step S401 input, in the specific implementation process, characteristics of image can be
But be not limited to: the grayscale information of image, the grayscale information of neighbor pixel, gradient information, neighbor pixel gradient information.It adopts
In aforementioned manners, convolutional neural networks model is trained, optimized, obtain third model, the third model is doubtful for classifying
Like localization region whether be target area two disaggregated models.It should be strongly noted that the third model belongs to two classification moulds
Type, wherein may be implemented using the third model: after the image input third model of doubtful localization region, judging doubtful positioning area
Whether target area is contained in the image in domain, and root exports the image in suspected target region in preset data label.
Step S405: the image of doubtful localization region is inputted into third model, obtains the image in suspected target region.
The third model that the image input step S403 of the doubtful localization region obtained abovementioned steps S103 is obtained, is used for
It detects whether containing suspected target region in doubtful positioning image, third model exports the image of doubtful positioning target.The classification
As a result it is two classification results, represents whether belong to target area.It should be strongly noted that the image in the suspected target region is
Cell area image.
Step S107: the 4th model is utilized, the image containing suspected target region is split, target area is obtained
Location information.
Abovementioned steps obtain the image containing suspected target region, need further exist for being split the image, to obtain
Obtain the location informations such as the location and shape of target area.Fig. 5 is the figure to suspected target region that this specification embodiment provides
As the flow chart being split, specifically include:
Step S501: input training sample to convolutional neural networks.
Sample for carrying out the 4th model training be the image of the zonules containing target area largely marked and
Training sample input convolutional neural networks are trained, the convolutional Neural net by the image of the zonule without containing target area
Network can be but be not limited to: the convolutional neural networks of 100 layer depths.It should be strongly noted that in order to guarantee training pattern
Accuracy is sufficiently large for trained sample size.
Step S503: fourth model of the training based on convolutional neural networks.
The characteristics of image of the training sample of extraction step S501 input, in the specific implementation process, characteristics of image can be
But be not limited to: the grayscale information of image, the grayscale information of neighbor pixel, gradient information, neighbor pixel gradient information.It adopts
In aforementioned manners, convolutional neural networks model is trained, optimized, obtain the 4th model.It can be real using the 4th model
It is existing: after the image in suspected target region inputs the 4th model, target area can be divided from the image of doubtful localization region
Out.
Step S505: the image in suspected target region is inputted into the 4th model, obtains the location information of target area.
In the 4th model that the image input step S503 in the suspected target region obtained abovementioned steps S105 is obtained, the
Four models are split the image in suspected target region, and target area is split, and obtain position and the shape of target area
The location informations such as shape.
The location information of target area: being reverted to the frame where target area by step S109, obtain target area to
Localization region in the two-dimentional DSA sequence image of processing.
The location information for the target area that abovementioned steps S107 is obtained belongs to the location information in the corresponding image in zonule,
Therefore, it is necessary to revert to the location information in the frame where target area.Specifically, location information based on target area
The location information of target area is reverted to the frame where target area, obtains target area by coordinate points by linear transformation
Corresponding coordinate points of the location information in the frame where target area;By the location information of target area where target area
The corresponding coordinate points of frame carry out interpolation line, obtain positioning area of the target area in two-dimentional DSA sequence image to be processed
Domain.In the specific implementation process, in order to guarantee the accuracy of target-region locating, when coordinate points are carried out line, coordinate points
Number is unsuitable very few, at least should include three coordinate points.Fig. 6 be this specification embodiment provide based on two-dimentional DSA sequence chart
The positioning result schematic diagram of picture, being capable of intuitive displaying target region.
The method provided using this specification embodiment, positions two-dimentional DSA sequence image, may be implemented in two dimension
Intuitive displaying target region in DSA sequence image excludes or reduces subjective factor and the diagnosis of image documentation equipment imaging difference bring
Difference reduces the time of artificial observation, thinking and judgement.
A kind of localization method based on DSA image is described in detail in above content, corresponding, present invention also provides
A kind of positioning system based on DSA image, as shown in Figure 7.Fig. 7 is that one kind that this specification embodiment provides is based on DSA image
Positioning system, specifically include:
Receiving unit 701 receives two-dimentional DSA sequence image to be processed;
Processing unit 703 positions two-dimentional DSA sequence image to be processed;
Output unit 705 shows the positioning result of two-dimentional DSA sequence image to be processed.
It is above-mentioned that this specification specific embodiment is described.Other embodiments are in the scope of the appended claims
It is interior.In some cases, the movement recorded in detail in the claims or step can be come according to the sequence being different from embodiment
It executes and desired result still may be implemented.In addition, process depicted in the drawing not necessarily require show it is specific suitable
Sequence or consecutive order are just able to achieve desired result.In some embodiments, multitasking and parallel processing be also can
With or may be advantageous.
All the embodiments in this specification are described in a progressive manner, same and similar portion between each embodiment
Dividing may refer to each other, and each embodiment focuses on the differences from other embodiments.Especially for device,
For electronic equipment, nonvolatile computer storage media embodiment, since it is substantially similar to the method embodiment, so description
It is fairly simple, the relevent part can refer to the partial explaination of embodiments of method.
Device that this specification embodiment provides, electronic equipment, nonvolatile computer storage media with method are corresponding
, therefore, device, electronic equipment, nonvolatile computer storage media also have the Advantageous effect similar with corresponding method
Fruit, since the advantageous effects of method being described in detail above, which is not described herein again corresponding intrument,
The advantageous effects of electronic equipment, nonvolatile computer storage media.
In the 1990s, the improvement of a technology can be distinguished clearly be on hardware improvement (for example,
Improvement to circuit structures such as diode, transistor, switches) or software on improvement (improvement for method flow).So
And with the development of technology, the improvement of current many method flows can be considered as directly improving for hardware circuit.
Designer nearly all obtains corresponding hardware circuit by the way that improved method flow to be programmed into hardware circuit.Cause
This, it cannot be said that the improvement of a method flow cannot be realized with hardware entities module.For example, programmable logic device
(Programmable Logic Device, PLD) (such as field programmable gate array (Field Programmable Gate
Array, FPGA)) it is exactly such a integrated circuit, logic function determines device programming by user.By designer
Voluntarily programming comes a digital display circuit " integrated " on a piece of PLD, designs and makes without asking chip maker
Dedicated IC chip.Moreover, nowadays, substitution manually makes IC chip, this programming is also used instead mostly " is patrolled
Volume compiler (logic compiler) " software realizes that software compiler used is similar when it writes with program development,
And the source code before compiling also write by handy specific programming language, this is referred to as hardware description language
(Hardware Description Language, HDL), and HDL is also not only a kind of, but there are many kind, such as ABEL
(Advanced Boolean Expression Language)、AHDL(Altera Hardware Description
Language)、Confluence、CUPL(Cornell University Programming Language)、HDCal、JHDL
(Java Hardware Description Language)、Lava、Lola、MyHDL、PALASM、RHDL(Ruby
Hardware Description Language) etc., VHDL (Very-High-Speed is most generally used at present
Integrated Circuit Hardware Description Language) and Verilog.Those skilled in the art also answer
This understands, it is only necessary to method flow slightly programming in logic and is programmed into integrated circuit with above-mentioned several hardware description languages,
The hardware circuit for realizing the logical method process can be readily available.
Controller can be implemented in any suitable manner, for example, controller can take such as microprocessor or processing
The computer for the computer readable program code (such as software or firmware) that device and storage can be executed by (micro-) processor can
Read medium, logic gate, switch, specific integrated circuit (Application Specific Integrated Circuit,
ASIC), the form of programmable logic controller (PLC) and insertion microcontroller, the example of controller includes but is not limited to following microcontroller
Device: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20 and Silicone Labs C8051F320 are deposited
Memory controller is also implemented as a part of the control logic of memory.It is also known in the art that in addition to
Pure computer readable program code mode is realized other than controller, can be made completely by the way that method and step is carried out programming in logic
Controller is obtained to come in fact in the form of logic gate, switch, specific integrated circuit, programmable logic controller (PLC) and insertion microcontroller etc.
Existing identical function.Therefore this controller is considered a kind of hardware component, and to including for realizing various in it
The device of function can also be considered as the structure in hardware component.Or even, it can will be regarded for realizing the device of various functions
For either the software module of implementation method can be the structure in hardware component again.
System, device, module or the unit that above-described embodiment illustrates can specifically realize by computer chip or entity,
Or it is realized by the product with certain function.It is a kind of typically to realize that equipment is computer.Specifically, computer for example may be used
Think personal computer, laptop computer, cellular phone, camera phone, smart phone, personal digital assistant, media play
It is any in device, navigation equipment, electronic mail equipment, game console, tablet computer, wearable device or these equipment
The combination of equipment.
For convenience of description, it is divided into various units when description apparatus above with function to describe respectively.Certainly, implementing this
The function of each unit can be realized in the same or multiple software and or hardware when specification one or more embodiment.
It should be understood by those skilled in the art that, this specification embodiment can provide as method, system or computer program
Product.Therefore, this specification embodiment can be used complete hardware embodiment, complete software embodiment or combine software and hardware
The form of the embodiment of aspect.Moreover, it wherein includes that computer is available that this specification embodiment, which can be used in one or more,
It is real in the computer-usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) of program code
The form for the computer program product applied.
This specification is referring to the method, equipment (system) and computer program product according to this specification embodiment
Flowchart and/or the block diagram describes.It should be understood that can be realized by computer program instructions every in flowchart and/or the block diagram
The combination of process and/or box in one process and/or box and flowchart and/or the block diagram.It can provide these computers
Processor of the program instruction to general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices
To generate a machine, so that generating use by the instruction that computer or the processor of other programmable data processing devices execute
In the dress for realizing the function of specifying in one or more flows of the flowchart and/or one or more blocks of the block diagram
It sets.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates,
Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or
The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or
The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one
The step of function of being specified in a box or multiple boxes.
In a typical configuration, calculating equipment includes one or more processors (CPU), input/output interface, net
Network interface and memory.
Memory may include the non-volatile memory in computer-readable medium, random access memory (RAM) and/or
The forms such as Nonvolatile memory, such as read-only memory (ROM) or flash memory (flash RAM).Memory is computer-readable medium
Example.
Computer-readable medium includes permanent and non-permanent, removable and non-removable media can be by any method
Or technology come realize information store.Information can be computer readable instructions, data structure, the module of program or other data.
The example of the storage medium of computer includes, but are not limited to phase change memory (PRAM), static random access memory (SRAM), moves
State random access memory (DRAM), other kinds of random access memory (RAM), read-only memory (ROM), electric erasable
Programmable read only memory (EEPROM), flash memory or other memory techniques, read-only disc read only memory (CD-ROM) (CD-ROM),
Digital versatile disc (DVD) or other optical storage, magnetic cassettes, tape magnetic disk storage or other magnetic storage devices
Or any other non-transmission medium, can be used for storage can be accessed by a computing device information.As defined in this article, it calculates
Machine readable medium does not include temporary computer readable media (transitory media), such as the data-signal and carrier wave of modulation.
It should also be noted that, the terms "include", "comprise" or its any other variant are intended to nonexcludability
It include so that the process, method, commodity or the equipment that include a series of elements not only include those elements, but also to wrap
Include other elements that are not explicitly listed, or further include for this process, method, commodity or equipment intrinsic want
Element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including described want
There is also other identical elements in the process, method of element, commodity or equipment.
This specification can describe in the general context of computer-executable instructions executed by a computer, such as journey
Sequence module.Generally, program module include routines performing specific tasks or implementing specific abstract data types, programs, objects,
Component, data structure etc..Specification can also be practiced in a distributed computing environment, in these distributed computing environments,
By executing task by the connected remote processing devices of communication network.In a distributed computing environment, program module can
To be located in the local and remote computer storage media including storage equipment.
All the embodiments in this specification are described in a progressive manner, same and similar portion between each embodiment
Dividing may refer to each other, and each embodiment focuses on the differences from other embodiments.Especially for system reality
For applying example, since it is substantially similar to the method embodiment, so being described relatively simple, related place is referring to embodiment of the method
Part explanation.
The foregoing is merely this specification embodiments, are not intended to limit this application.For those skilled in the art
For, various changes and changes are possible in this application.All any modifications made within the spirit and principles of the present application are equal
Replacement, improvement etc., should be included within the scope of the claims of this application.
Claims (15)
1. a kind of localization method based on DSA image, which is characterized in that the described method includes:
Using the first model, from two-dimentional DSA sequence image to be processed, the frame containing target area is obtained, wherein described the
One model is the model being obtained ahead of time based on deep learning method, and the two-dimentional DSA sequence image to be processed is multiframe;
Based on the frame containing target area, the doubtful localization region of target area is extracted, wherein the doubtful localization region
It is made of a series of doubtful zonules containing target area;
Using third model, classify to the image of the doubtful localization region, obtain the image in suspected target region,
In, the third model is two disaggregated models being obtained ahead of time based on deep learning method, and the suspected target region is to be based on
The possible target area of the third model output;
Using the 4th model, the image containing the suspected target region is split, the location information of target area is obtained,
Wherein, the 4th model is the model being obtained ahead of time based on deep learning method;
The location information of the target area is reverted to the frame where target area, obtains target area described to be processed
Localization region in two-dimentional DSA sequence image.
2. the method as described in claim 1, which is characterized in that the method also includes pre-treatment steps:
The two-dimentional DSA sequence image to be processed is normalized, image transformation, scaling, interception and filling processing
One of or a variety of processing, make the picture size and picture of the two-dimentional DSA sequence image to be processed and first model
Plain spacing is consistent.
3. the method as described in claim 1, which is characterized in that it is described to utilize the first model, from two-dimentional DSA sequence to be processed
In image, the frame containing target area is obtained, is specifically included:
The two-dimentional DSA sequence image to be processed is inputted into the first model, obtains in the two dimension DSA sequence image and contains mesh
Mark the frame in region.
4. the method as described in claim 1, which is characterized in that it is described based on the frame containing target area, extract target
The doubtful localization region in region, specifically includes:
Using the second model, from the frame containing target area, the corresponding temperature of the frame containing target area is determined
Figure;
According to the second preset threshold, each pixel for belonging to target area in the temperature figure is determined;
Based on the coordinate for each pixel for belonging to target area in the temperature figure, intercepted from the frame containing target area
The corresponding region of the coordinate, obtains the doubtful localization region of target area.
5. the method as described in claim 1, which is characterized in that it is described to utilize third model, to the doubtful localization region
Image is classified, and is obtained the image in suspected target region, is specifically included:
The third model judges whether the doubtful localization region is target according to the characteristics of image of the doubtful localization region
Region exports the image in suspected target region.
6. the method as described in claim 1, which is characterized in that it is described utilize the 4th model, to contain the suspected target area
The image in domain is split, and is obtained the location information of target area, is specifically included:
The image in the suspected target region is inputted the 4th model to be split, the 4th model output contains target
The image in region and corresponding coordinate information, obtain the location information of target area.
7. the method as described in claim 1, which is characterized in that the location information by the target area reverts to target
Frame where region obtains localization region of the target area in the two-dimentional DSA sequence image to be processed, specifically includes:
The coordinate points of location information based on the target area, by linear transformation, by the location information of the target area
The frame where the target area is reverted to, obtains the location information of the target area in the frame where the target area
Corresponding coordinate points;
Corresponding coordinate points by the location information of the target area in the frame where the target area carry out interpolation line,
Obtain localization region of the target area in the two-dimentional DSA sequence image to be processed.
8. a kind of positioning system based on DSA image, which is characterized in that the system comprises:
Receiving unit receives two-dimentional DSA sequence image to be processed;
Processing unit positions the two-dimentional DSA sequence image to be processed;
Output unit shows the positioning result of the two-dimentional DSA sequence image to be processed.
9. system as claimed in claim 8, described to position to the two-dimentional DSA sequence image to be processed, specific to wrap
It includes:
Using the first model, from two-dimentional DSA sequence image to be processed, the frame containing target area is obtained, wherein described the
One model is the model being obtained ahead of time based on deep learning method, and the two-dimentional DSA sequence image to be processed is multiframe;
Based on the frame containing target area, the doubtful localization region of target area is extracted, wherein the doubtful localization region
It is made of a series of doubtful zonules containing target area;
Using third model, classify to the image of the doubtful localization region, obtains suspected target region, wherein described
Third model is two disaggregated models being obtained ahead of time based on deep learning method, and the suspected target region is based on the third
The possible target area of model output;
Using the 4th model, the image containing the suspected target region is split, the location information of target area is obtained,
Wherein, the 4th model is the model being obtained ahead of time based on deep learning method;
The location information of the target area is reverted to the frame where target area, obtains target area described to be processed
Localization region in two-dimentional DSA sequence image.
10. system as claimed in claim 9, which is characterized in that the method also includes pre-treatment steps:
The two-dimentional DSA sequence image to be processed is normalized, image transformation, scaling, interception and filling processing
One of or a variety of processing, make the picture size and picture of the two-dimentional DSA sequence image to be processed and first model
Plain spacing is consistent.
11. system as claimed in claim 9, which is characterized in that it is described to utilize the first model, from two-dimentional DSA sequence to be processed
In column image, the frame containing target area is obtained, is specifically included:
The two-dimentional DSA sequence image to be processed is inputted into the first model, obtains in the two dimension DSA sequence image and contains mesh
Mark the frame in region.
12. system as claimed in claim 9, which is characterized in that it is described based on the frame containing target area, extract target
The doubtful localization region in region, specifically includes:
Using the second model, from the frame containing target area, the corresponding temperature of the frame containing target area is determined
Figure;
According to the second preset threshold, each pixel for belonging to target area in the temperature figure is determined;
Based on the coordinate for each pixel for belonging to target area in the temperature figure, intercepted from the frame containing target area
The corresponding region of the coordinate, obtains the doubtful localization region of target area.
13. system as claimed in claim 9, which is characterized in that it is described to utilize third model, to the doubtful localization region
Image is classified, and is obtained the image in suspected target region, is specifically included:
The third model judges whether the doubtful localization region is target according to the characteristics of image of the doubtful localization region
Region exports the image in suspected target region.
14. system as claimed in claim 9, which is characterized in that it is described utilize the 4th model, to contain the suspected target area
The image in domain is split, and is obtained the location information of target area, is specifically included:
The image in the suspected target region is inputted the 4th model to be split, the 4th model output contains target
The image in region and corresponding coordinate information, obtain the location information of target area.
15. system as claimed in claim 9, which is characterized in that the location information by the target area reverts to mesh
The frame where region is marked, localization region of the target area in the two-dimentional DSA sequence image to be processed is obtained, it is specific to wrap
It includes:
The coordinate points of location information based on the target area, by linear transformation, by the location information of the target area
The frame where the target area is reverted to, obtains the location information of the target area in the frame where the target area
Corresponding coordinate points;
Corresponding coordinate points by the location information of the target area in the frame where the target area carry out interpolation line,
Obtain localization region of the target area in the two-dimentional DSA sequence image to be processed.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910782514.XA CN110517243B (en) | 2019-08-23 | 2019-08-23 | Positioning method and system based on DSA image |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910782514.XA CN110517243B (en) | 2019-08-23 | 2019-08-23 | Positioning method and system based on DSA image |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110517243A true CN110517243A (en) | 2019-11-29 |
CN110517243B CN110517243B (en) | 2022-03-25 |
Family
ID=68627248
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910782514.XA Active CN110517243B (en) | 2019-08-23 | 2019-08-23 | Positioning method and system based on DSA image |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110517243B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113011509A (en) * | 2021-03-25 | 2021-06-22 | 推想医疗科技股份有限公司 | Lung bronchus classification method and device, electronic equipment and storage medium |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107578416A (en) * | 2017-09-11 | 2018-01-12 | 武汉大学 | It is a kind of by slightly to heart left ventricle's full-automatic partition method of smart cascade deep network |
US9934364B1 (en) * | 2017-02-28 | 2018-04-03 | Anixa Diagnostics Corporation | Methods for using artificial neural network analysis on flow cytometry data for cancer diagnosis |
CN107945181A (en) * | 2017-12-30 | 2018-04-20 | 北京羽医甘蓝信息技术有限公司 | Treating method and apparatus for breast cancer Lymph Node Metastasis pathological image |
CN109801276A (en) * | 2019-01-14 | 2019-05-24 | 沈阳联氪云影科技有限公司 | A kind of method and device calculating ambition ratio |
CN110070546A (en) * | 2019-04-18 | 2019-07-30 | 山东师范大学 | A kind of multiple target based on deep learning jeopardizes the automatic division method of organ, apparatus and system |
-
2019
- 2019-08-23 CN CN201910782514.XA patent/CN110517243B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9934364B1 (en) * | 2017-02-28 | 2018-04-03 | Anixa Diagnostics Corporation | Methods for using artificial neural network analysis on flow cytometry data for cancer diagnosis |
CN107578416A (en) * | 2017-09-11 | 2018-01-12 | 武汉大学 | It is a kind of by slightly to heart left ventricle's full-automatic partition method of smart cascade deep network |
CN107945181A (en) * | 2017-12-30 | 2018-04-20 | 北京羽医甘蓝信息技术有限公司 | Treating method and apparatus for breast cancer Lymph Node Metastasis pathological image |
CN109801276A (en) * | 2019-01-14 | 2019-05-24 | 沈阳联氪云影科技有限公司 | A kind of method and device calculating ambition ratio |
CN110070546A (en) * | 2019-04-18 | 2019-07-30 | 山东师范大学 | A kind of multiple target based on deep learning jeopardizes the automatic division method of organ, apparatus and system |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113011509A (en) * | 2021-03-25 | 2021-06-22 | 推想医疗科技股份有限公司 | Lung bronchus classification method and device, electronic equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN110517243B (en) | 2022-03-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109493348A (en) | A kind of measurement method and system of the Morphologic Parameters of intracranial aneurysm image | |
CN109345585A (en) | A kind of measurement method and system of the Morphologic Parameters of intracranial aneurysm image | |
CN110517780A (en) | A kind of aneurysm rupture methods of risk assessment and system | |
CN109658455A (en) | Image processing method and processing equipment | |
CN109584997B (en) | Method and system for measuring morphological parameters of intracranial aneurysm image | |
CN109448003A (en) | A kind of entocranial artery blood-vessel image dividing method and system | |
CN109658400A (en) | A kind of methods of marking and system based on head CT images | |
CN109389637A (en) | A kind of measurement method and system of the Morphologic Parameters of intracranial aneurysm image | |
CN107977634A (en) | A kind of expression recognition method, device and equipment for video | |
CN109472780A (en) | A kind of measurement method and system of the Morphologic Parameters of intracranial aneurysm image | |
CN109472823B (en) | Method and system for measuring morphological parameters of intracranial aneurysm image | |
CN110534193A (en) | A kind of aneurysm rupture methods of risk assessment and system | |
WO2020118618A1 (en) | Mammary gland mass image recognition method and device | |
CN109671066A (en) | A kind of method and system of the cerebral infarction judgement based on head CT images | |
CN110517242A (en) | A kind of aneurysmal analysis method and device | |
CN109584169A (en) | A kind of intercept method and system of the intracranial vessel image based on center line | |
CN109448004A (en) | A kind of intercept method and system of the intracranial vessel image based on center line | |
CN116187448B (en) | Information display method and device, storage medium and electronic equipment | |
CN109658399A (en) | A kind of neck patch image-recognizing method and device | |
CN109447967A (en) | A kind of dividing method and system of intracranial aneurysm image | |
CN110517243A (en) | A kind of localization method and system based on DSA image | |
CN110503642A (en) | A kind of localization method and system based on DSA image | |
CN109712122A (en) | A kind of methods of marking and system based on head CT images | |
CN110517244A (en) | A kind of localization method and system based on DSA image | |
CN109671067A (en) | A kind of measurement method and system of the core Infarction volume based on head CT images |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |