CN109345546B - A kind of coronary artery volume data model dividing method and equipment - Google Patents
A kind of coronary artery volume data model dividing method and equipment Download PDFInfo
- Publication number
- CN109345546B CN109345546B CN201811162108.5A CN201811162108A CN109345546B CN 109345546 B CN109345546 B CN 109345546B CN 201811162108 A CN201811162108 A CN 201811162108A CN 109345546 B CN109345546 B CN 109345546B
- Authority
- CN
- China
- Prior art keywords
- segmentation
- information
- segmentation result
- name
- coronary artery
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 210000004351 coronary vessel Anatomy 0.000 title claims abstract description 85
- 238000013499 data model Methods 0.000 title claims abstract description 66
- 238000000034 method Methods 0.000 title claims abstract description 34
- 230000011218 segmentation Effects 0.000 claims abstract description 337
- 210000003462 vein Anatomy 0.000 claims abstract description 63
- 210000004204 blood vessel Anatomy 0.000 claims description 24
- 230000008439 repair process Effects 0.000 claims description 19
- 235000013399 edible fruits Nutrition 0.000 claims description 8
- 238000000605 extraction Methods 0.000 claims description 5
- 238000003745 diagnosis Methods 0.000 abstract description 5
- 239000003814 drug Substances 0.000 abstract description 5
- 230000002159 abnormal effect Effects 0.000 description 17
- 238000001514 detection method Methods 0.000 description 5
- 239000000284 extract Substances 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 230000002526 effect on cardiovascular system Effects 0.000 description 3
- 210000003484 anatomy Anatomy 0.000 description 2
- 210000001367 artery Anatomy 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 239000008280 blood Substances 0.000 description 1
- 210000004369 blood Anatomy 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000002591 computed tomography Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000002059 diagnostic imaging Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 230000002107 myocardial effect Effects 0.000 description 1
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
-
- G06T5/77—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10081—Computed x-ray tomography [CT]
-
- 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 embodiment of the invention discloses a kind of coronary artery volume data model dividing method and coronary artery volume data model splitting equipments, which comprises is split to coronary artery volume data model, obtains including the first segmentation result for naming information；According to name list to include name information the first segmentation result present in segmentation risk position position；Oriented segmentation risk position is repaired, the second segmentation result after being repaired；The second segmentation result after exporting the reparation.The present invention is by carrying out risk supervision to coronary artery volume data model primary segmentation result, there may be the positions of segmentation risk for positioning, reparation improved model is carried out for the segmentation risk detected again, the problems such as can effectively solve the problem that omission present in the segmentation of coronary artery volume data model, fracture and vein noise, and then automation coronary artery model accuracy is improved, to provide more valuable information for clinical medicine diagnosis.
Description
Technical field
The present invention relates to Medical Imaging Technology field, in particular to a kind of coronary artery volume data model dividing method and equipment.
Background technique
In modern medicine image analysis, there is important clinical value to doctor to automation coronary artery model reconstruction techniques
And practical significance, automation coronary artery Model Reconstruction need to use medical image cutting method.Medical image segmentation, which just refers to, to be schemed
Clinically disease is examined to be handled medical image and be analyzed as the basic theories of segmentation is applied to medical image
Disconnected and treatment provides help.
But in practical situations, since medical image has the characteristics that high complexity, coronary artery model divides automatically
It is often influenced by vein when cutting, and segmentation can occur and lose because of myocardial bridge, staggered floor artifact or due to the defect of parted pattern due to
The problems such as leakage, fracture, vein noise, therefore, it is necessary to further repair to segmentation result.
Summary of the invention
The embodiment of the present invention is existing in order to effectively solve the problems, such as, creatively provides a kind of coronary artery volume data model
Dividing method and equipment.
According to the first aspect of the invention, a kind of coronary artery volume data model dividing method is provided, which comprises
Coronary artery volume data model is split, obtains including the first segmentation result for naming information；It is arranged according to name
Table to include name information the first segmentation result present in segmentation risk position position；To oriented segmentation wind
Dangerous position is repaired, the second segmentation result after being repaired；The second segmentation result after exporting the reparation.
In one embodiment, according to the above method of the present invention, coronary artery volume data model is split, including
The first segmentation result for having name information includes: during being split to coronary artery volume data model, to coronary artery volume data
Model is classified, and multiple category informations are obtained；To each category information Corresponding matching one in the multiple category information
The name information of branch, so that obtaining includes the first segmentation result for naming information；Or, being carried out to coronary artery volume data model
During segmentation, extraction obtains center line；The center line and segmentation result are named automatically, obtain including name
First segmentation result of information.
In one embodiment, according to the above method of the present invention, it is described according to name list to including name information
The first segmentation result present in segmentation risk position positioned, comprising: information will be named in first segmentation result
It is matched with the fixed name in the name list；If not found in first segmentation result and the first fixed name
The the first name information matched, then positioning and corresponding to the position of the first name information in first segmentation result is segmentation risk
Position.
In one embodiment, according to the above method of the present invention, it is described according to name list to including name information
The first segmentation result present in segmentation risk position positioned, comprising: each life in detection first segmentation result
The data length of name information；Data length is less than the second name information of length threshold if it exists, then first segmentation result
The position of middle correspondence the second name information is segmentation risk position.
In one embodiment, according to the above method of the present invention, it is described according to name list to including name information
The first segmentation result present in segmentation risk position positioned, comprising: information will be named in first segmentation result
In location information with it is described name list in location information matched, if in first segmentation result name information pair
It answers the location information in first location information and the name list to mismatch, then positions in first segmentation result and correspond to institute
The position of first location information is stated as segmentation risk position；Or, information and the name will be named in first segmentation result
Fixed name in list is matched, and names unmatched third name to believe with fixed if finding in first segmentation result
Breath, then positioning and corresponding to the position of the third name information in first segmentation result is segmentation risk position.
In one embodiment, according to the above method of the present invention, described that oriented segmentation risk position is repaired
It is multiple, the segmentation result after being repaired, comprising: the segmentation risk position in first segmentation result is to coronary artery volume data mould
Type is matched by adjusting model segmentation confidence level, obtains the first name result；If the first name result meets described
The first name result correspondence is then inserted first segmentation result by the first fixed name.
In one embodiment, according to the above method of the present invention, described that oriented segmentation risk position is repaired
It is multiple, the segmentation result after being repaired, comprising: to the segmentation risk place-centric line extending direction in first segmentation result
Data matched, obtain the first dividing body；If the data length of first dividing body is the second name information
The difference of data length and the length threshold then by algorithm or adjusts confidence level for described first point in segmentation risk position
Body is cut to be attached with first segmentation result.
In one embodiment, according to the above method of the present invention, described that oriented segmentation risk position is repaired
It is multiple, the segmentation result after being repaired, comprising: vein is carried out to the segmentation risk position data of first segmentation result and is sentenced
It is fixed, it obtains vein and determines result；If the vein determine result be it is yes, risk will be divided described in first segmentation result
Venous information removal at position；Wherein, the vein determines to include such as at least one under type: confidence level is adjusted, and vein is known
Do not judge with vein tendency.
Another aspect of the present invention provides a kind of coronary artery volume data model splitting equipment, and the equipment includes:
Divide module, for dividing coronary artery volume data model, obtains including the first segmentation result for naming information；Positioning
Module, for according to name list to include name information the first segmentation result present in segmentation risk position determine
Position；Repair module, for being repaired to oriented segmentation risk position, the second segmentation result after being repaired；Output
Module, for exporting the second segmentation result after the reparation.
In one embodiment, above equipment according to the present invention, the segmentation module further include: classification cutting unit,
For classifying to coronary artery volume data model, obtaining multiple classifications during being split to coronary artery volume data model
Information；To the name information of one branch of each category information Corresponding matching in the multiple category information, to be wrapped
Include the first segmentation result of name information；Or, center line cutting unit, for being split to coronary artery volume data model
In the process, it extracts and obtains center line；The center line and segmentation result are named automatically, obtain including name information
First segmentation result.
In one embodiment, above equipment according to the present invention, the locating module include: name matching unit, are used
It is matched in information will be named to name in first segmentation result with the fixation in the name list；First positioning is single
Member, if matched first name information is named for not finding to fix with first in first segmentation result, described in positioning
The position of the first name information is corresponded in first segmentation result as segmentation risk position.
In one embodiment, above equipment according to the present invention, the locating module include: that data length detection is single
Member, for detecting the data length of each name information in first segmentation result；Second positioning unit, for counting if it exists
It is less than the second name information of length threshold according to length, then corresponds to the position of the second name information in first segmentation result
It sets and is positioned as dividing risk position.
In one embodiment, above equipment according to the present invention, the locating module include: location matches unit, are used
It is matched in by the location information in first segmentation result in name information with the location information in the name list；
Third positioning unit, if being corresponded in first location information and the name list for name information in first segmentation result
Location information mismatch, then position correspond in first segmentation result first location information position be divide risk
Position；Further include, it is abnormal to name matching unit, for will be named in information and the name list in first segmentation result
Fixed name matched；4th positioning unit, if being mismatched for being found in first segmentation result with fixed name
Third name information, then positioning and correspond to the position of third name information in first segmentation result is to divide risk position
It sets.
In one embodiment, above equipment according to the present invention, the repair module include: adjusting matching unit, are used
In the segmentation risk position in first segmentation result to coronary artery volume data model by adjust model divide confidence level into
Row matching, obtains the first name result；First repairs unit, if meeting the described first fixed life for the first name result
The first name result correspondence is then inserted first segmentation result by name.
In one embodiment, above equipment according to the present invention, the repair module include: data matching unit, are used
It is matched in the data to the segmentation risk place-centric line extending direction in first segmentation result, obtains the first segmentation
Body；Second repair unit, if for first dividing body data length be it is described second name information data length with
The difference of the length threshold, then segmentation risk position by algorithm or adjust confidence level by first dividing body with it is described
First segmentation result is attached.
In one embodiment, above equipment according to the present invention, the repair module include: vein judging unit, are used
Vein judgement is carried out in the segmentation risk position data to first segmentation result, vein is obtained and determines result；Third reparation
Unit, if for the vein determine result be it is yes, will described in first segmentation result divide risk position at it is quiet
The removal of arteries and veins information；Wherein, the vein determines to include such as at least one under type: confidence level is adjusted, and hand vein recognition and vein are walked
Gesture judgement.
Coronary artery volume data model dividing method of the present invention and equipment, by coronary artery volume data model primary segmentation knot
Fruit carries out risk supervision, and there may be the positions of segmentation risk for positioning, then carry out reparation improvement for the segmentation risk detected
Model, can effectively solve the problem that omitted present in the segmentation of coronary artery volume data model, fracture and the problems such as vein noise, and then improve
Coronary artery model accuracy is automated, to provide more valuable information for clinical medicine diagnosis.
Detailed description of the invention
The following detailed description is read with reference to the accompanying drawings, above-mentioned and other mesh of exemplary embodiment of the invention
, feature and advantage will become prone to understand.In the accompanying drawings, if showing by way of example rather than limitation of the invention
Dry embodiment, in which:
In the accompanying drawings, identical or corresponding label indicates identical or corresponding part.
Fig. 1 is the implementation process schematic diagram of coronary artery dividing method of the present invention；
Fig. 2 is the structure chart of coronary artery volume data model splitting equipment embodiment of the present invention.
Specific embodiment
To keep the purpose of the present invention, feature, advantage more obvious and understandable, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is only
It is only a part of the embodiment of the present invention, and not all embodiments.Based on the embodiments of the present invention, those skilled in the art are not having
Every other embodiment obtained under the premise of creative work is made, shall fall within the protection scope of the present invention.
In the description of this specification, reference term " one embodiment ", " some embodiments ", " example ", " specifically show
The description of example " or " some examples " etc. means specific features, structure, material or spy described in conjunction with this embodiment or example
Point is included at least one embodiment or example of the invention.Moreover, particular features, structures, materials, or characteristics described
It may be combined in any suitable manner in any one or more of the embodiments or examples.In addition, without conflicting with each other, this
The technical staff in field can be by the spy of different embodiments or examples described in this specification and different embodiments or examples
Sign is combined.
In addition, term " first ", " second " are used for descriptive purposes only and cannot be understood as indicating or suggesting relative importance
Or implicitly indicate the quantity of indicated technical characteristic." first " is defined as a result, the feature of " second " can be expressed or hidden
It include at least one this feature containing ground.In the description of the present invention, the meaning of " plurality " is two or more, unless otherwise
Clear specific restriction.
With reference to Fig. 1, Fig. 1 is the implementation process schematic diagram of coronary artery dividing method of the present invention.
One aspect of the present invention provides a kind of coronary artery volume data model dividing method, this method comprises:
Step 101, coronary artery volume data model is split, obtains including the first segmentation result for naming information；
Wherein, coronary artery volume data model is modeled according to multi-Slice CT figure, and model, which includes at least, each blood vessel of coronary artery
Three dimensional local information, blood vessels caliber information and blood vessel move towards information.It is obtained after being split to three-dimensional coronary artery volume data model
It include the first segmentation result for naming information, the first segmentation result includes multiple segmentation the data obtained groups, each data group
A corresponding name information is matched respectively, wherein name information, which includes at least, has three dimensional local information, length information, blood vessel to name,
Blood vessels caliber information and blood vessel move towards information.
Step 102, according to name list to include name information the first segmentation result present in segmentation risk position
It sets and is positioned；
Wherein, name list is according to international cardiovascular CT association SCCT (Society of Cardiovascular
Computed Tomography) cardiovascular naming standard, likewise, name list not only include being built according to SCCT
Vertical fixed name set, also includes at least the corresponding location information of coronary artery body blood vessel, length information, blood vessels caliber information and blood
Pipe moves towards information.Name information in first segmentation result is usually document form data, and filename is coronary artery name, such as
LAD (left anterior descending branch), LCX (Zuo Huixuan), the data in file are the three-dimensional of each coronary branches blood vessel in the first segmentation result
Location information, length information, blood vessels caliber information and blood vessel move towards information data.
In the present embodiment, believed by the fixed name or length of naming information in comparison name list and the first segmentation result
Breath etc. tentatively can determine whether out the data group of abnormal position, and the three dimensional local information extracted in data group can be in coronary artery volume data mould
Segmentation risk position is oriented in matching in type.
Step 103, oriented segmentation risk position is repaired, the second segmentation result after being repaired；
Step 104, the second segmentation result after output is repaired.
In the present embodiment, coronary artery volume data model primary segmentation result is detected by comparison name list, is obtained
There may be the positions of segmentation risk, then carry out repairing improvement mould for the segmentation risk problem detected at segmentation risk position
Type, can effectively solve the problem that omitted present in the segmentation of coronary artery volume data model, fracture and the problems such as vein noise, and then improve from
Dynamicization coronary artery model accuracy, to provide more valuable information for clinical medicine diagnosis.
In the present embodiment, classifies to coronary artery volume data model, obtain multiple category informations；To in multiple category informations
The name information of one branch of each category information Corresponding matching, so that obtaining includes the first segmentation knot for naming information
Fruit；Or, extraction obtains center line during being split to coronary artery volume data model；To center line and segmentation result into
The automatic name of row, obtains including the first segmentation result for naming information.
Wherein, it carries out classification to coronary artery volume data model to be specifically as follows, according to the three-dimensional position of coronary artery blood vessel, blood vessel is long
Degree and blood vessels caliber, move towards information for it and are categorized into data group one by one according to SCCT anatomical features, obtain comprising just like a left side
Before be down to, the data folder of the specific category information of arteria coronaria dextra, then according to SCCT naming standard correspondence name one by one
The data obtained file, so that obtaining includes the first segmentation result for naming information.
Or using during being split to coronary artery volume data model, identification extracts and obtains vessel centerline, then root
Multiple data groups are divided into according to center line and related data, obtain multiple data files comprising center line and its related data
Folder, and then multiple data folders are named automatically also according to SCCT naming standard, so that obtaining includes name letter
First segmentation result of breath.
In one embodiment, according to name list to include name information the first segmentation result present in divide wind
It includes: that information will be named to match in the first segmentation result with the fixed name in list is named that dangerous position, which carries out positioning,；If
It is not found in first segmentation result and names information with the first fixed name matched first, then positioned corresponding in the first segmentation result
The position of first name information is segmentation risk position.
Wherein, information will be named to be matched with the fixed name in name list in the first segmentation result specially to compare
The fixed name in filename and name list in first segmentation result, if not found in the first segmentation result and name list
In the corresponding first name information of the first fixed name, omit it would be possible that there is segmentation, searched from the first segmentation result
The data segment that do not named is positioned as the data segment to divide risk position to be quickly and accurately positioned out and divides the position of omission
It sets.
In another embodiment, according to name list to include name information the first segmentation result present in segmentation
Risk position is positioned, comprising: the data length of each name information in the first segmentation result of detection；Data length if it exists
Second less than length threshold names information, then the position of corresponding second name information is segmentation risk position in the first segmentation result
It sets.
Wherein, length threshold is the controller perturbation range being arranged according to each length of vessel of SCCT anatomy, when first point
The length threshold that the data group data length in result is less than setting is cut, then there may be segmentations to be broken for explanation, then extracting small
Location information in the data group of length threshold, and the position is positioned in the first segmentation result as segmentation risk position.
In another embodiment, according to name list to include name information the first segmentation result present in segmentation
Risk position is positioned, comprising: by the position in the location information and name list named in information in the first segmentation result
Information is matched, if information is named to correspond to the location information in first location information and name list not in the first segmentation result
Matching, then positioning and corresponding to the position of first location information in the first segmentation result is segmentation risk position；Or, the first segmentation is tied
It names information to be matched in fruit with the fixed name in list is named, is not named not if being found in the first segmentation result with fixed
The third name information matched, then positioning and corresponding to the position of third name information in the first segmentation result is segmentation risk position.
Wherein, the location information in the first segmentation result is matched with the location information in name list, if first
There is the location information in the location information or the first segmentation result being not present in name list and name list in segmentation result
Middle location information has very important error, then there may be veins for explanation, then being tied according to the location information in the first segmentation
Locating segmentation risk position in fruit.Or, being ordered by fixing in the name information and name list in the first segmentation result of comparison
Name exists and fixation in unexistent fixed name or the first segmentation result if existing in name list in the first segmentation result
Name none matched name, then there may be veins for explanation, then the location information in abnormal name information data group is extracted,
And the position is positioned as in the first segmentation result to divide risk position.
The position that may be present for dividing omission, segmentation fracture and vein noise in segmentation is oriented through the above steps
Afterwards, it needs that further it is judged and is repaired, thus the second segmentation result after being repaired.
In one embodiment, oriented segmentation risk position is repaired, the segmentation result after being repaired, is wrapped
It includes: being divided by adjusting model to coronary artery volume data model by confidence level progress for the segmentation risk position in the first segmentation result
Match, obtains the first name result；If the first name result meets the first fixed name, the is inserted by the first name result is corresponding
One segmentation result.
Wherein, set trust threshold when model segmentation confidence level is to model segmentation, in the first segmentation result
Divide risk position and adjust confidence level, such as can attempt know the region again access evidence by turning down confidence level, obtain corresponding to the
One name result and its corresponding data group, if the first name result meets the first fixed name of above-mentioned omission, by the first life
Name result and its corresponding data group insert the first segmentation result.
In another embodiment, oriented segmentation risk position is repaired, the segmentation result after being repaired, is wrapped
It includes: the data of the segmentation risk place-centric line extending direction in the first segmentation result being matched, the first dividing body is obtained；
If the data length of the first dividing body is the data length of the second name information and the difference of length threshold, in segmentation risk position
It sets and is attached the first dividing body and the first segmentation result by algorithm or adjusting confidence level.
Wherein, the data of the segmentation risk place-centric line extending direction in the first segmentation result are matched to obtain
One dividing body specifically, knowing the data segment for taking above-mentioned segmentation risk position, and reads its position data and obtains segmentation risk position
Center line extending direction data, in the data in this direction further know take the first dividing body, first dividing body is three
It ties up location information, length of vessel, caliber and moves towards to match with segmentation risk position data in information.After obtaining the first dividing body
Further the data length of the first dividing body is identified, if the data length be just the second name information data length with
The difference of length threshold then illustrates to know obtaining the first dividing body as required segmentation fracture location data, leads in segmentation risk position
It crosses algorithm or the first dividing body and the first segmentation result is attached by confidence level adjusting.
In another embodiment, oriented segmentation risk position is repaired, the segmentation result after being repaired, is wrapped
It includes: vein judgement being carried out to the segmentation risk position data of the first segmentation result, vein is obtained and determines result；If vein determines knot
Fruit be it is yes, then by the first segmentation result divide risk position at venous information remove；Wherein, vein determines to include such as lower section
At least one formula: confidence level is adjusted, hand vein recognition and the judgement of vein tendency.
Wherein, vein judgement is carried out to the segmentation risk position of the first segmentation result, the mode that vein determines includes: confidence
Whether degree is adjusted, can specifically be disappeared with abnormal name situation to determine whether being quiet by the way that confidence level observation abnormal position is turned up
Arteries and veins；Hand vein recognition, specially to three dimensional local information, blood vessels caliber information in abnormal name or abnormal position data group etc. into
Row identification judges whether it is vein；It further include the judgement of vein tendency, specially in abnormal name or abnormal position data group
Blood vessel moves towards information and carries out judging whether to meet vein trend characteristic.If the determination result is YES, then by the first segmentation result points
Cut the venous information removal at risk position, the second segmentation result after being repaired.
With reference to Fig. 2, Fig. 2 is the structure chart of coronary artery volume data model splitting equipment embodiment of the present invention.
On the other hand, the present invention also provides a kind of coronary artery volume data model splitting equipment, the segmentation of coronary artery volume data model is set
Standby includes: segmentation module 201, for dividing coronary artery volume data model, obtains including the first segmentation result for naming information；It is fixed
Position module 202, for according to name list to include name information the first segmentation result present in segmentation risk position
It is positioned；Repair module 203, the second segmentation for being repaired to oriented segmentation risk position, after being repaired
As a result；Output module 204, for exporting the second segmentation result after repairing.
Wherein, the three-dimensional coronary artery volume data models of 201 pairs of module of segmentation obtain including name information the after being split
One segmentation result, the first segmentation result include multiple segmentation the data obtained groups, and each data group matches a corresponding name respectively
Information.Fixed name or length information of the locating module 202 by name information in comparison name list and the first segmentation result
Deng the preliminary data group that can determine whether out abnormal position, the three dimensional local information extracted in data group can be in coronary artery volume data model
Segmentation risk position is oriented in middle matching.
Divide module 201 in the present embodiment to obtaining the first segmentation result after coronary artery volume data model primary segmentation, then leads to
It crosses locating module 202 and 203 precise positioning of repair module and repairs segmentation risk, improve segmentation result, efficiently solve coronary artery
The problems such as omission present in the segmentation of volume data model, fracture and vein noise, and then automation coronary artery model accuracy is improved,
More valuable information is provided for clinical medicine diagnosis.
In the present embodiment, divide module 201 further include: classification cutting unit, for being carried out to coronary artery volume data model
During segmentation, classifies to coronary artery volume data model, obtain multiple category informations；To each in multiple category informations
The name information of one branch of category information Corresponding matching, so that obtaining includes the first segmentation result for naming information；In or,
Heart line cutting unit, for during being split to coronary artery volume data model, extraction to obtain center line；To center line and
Segmentation result is named automatically, obtains including the first segmentation result for naming information.
Wherein, for classification cutting unit according to the three-dimensional position of coronary artery blood vessel, length of vessel and blood vessels caliber, moving towards information will
It is categorized into data group according to SCCT anatomical features one by one, is obtained comprising being down to just like left front, arteria coronaria dextra it is specific
Then the data folder of category information names the data obtained file according to SCCT naming standard is corresponding, to obtain one by one
It include the first segmentation result for naming information；And the process that center line cutting unit is split coronary artery volume data model
In, identification extracts and obtains vessel centerline, is then divided into multiple data groups according to center line and related data, is included
Multiple data folders of heart line and its related data, and then multiple data folders are carried out also according to SCCT naming standard
Automatic name, so that obtaining includes the first segmentation result for naming information.
In the present embodiment, locating module 202 includes: name matching unit, for will name information in the first segmentation result
It is matched with the fixed name in name list；First positioning unit, if for not found in the first segmentation result and first
The fixed matched first name information of name, then position the position of corresponding first name information in the first segmentation result as segmentation wind
Dangerous position.
Wherein, name matching unit is specifically used for the fixation in the filename in the first segmentation result of comparison and name list
Name names information with the first fixed name corresponding first in name list if not finding in the first segmentation result, then
There may be segmentations to omit；First positioning unit searches the data segment that do not named from the first segmentation result, by the data segment
The position for dividing omission can be quickly and accurately positioned out by being positioned as segmentation risk position.
In another embodiment, locating module 202 includes: data length detection unit, for detecting the first segmentation result
In it is each name information data length；Second positioning unit is less than the second life of length threshold for data length if it exists
Name information, then the position of corresponding second name information is positioned as dividing risk position in the first segmentation result.
Wherein, whether the data group data length that data detecting unit is specifically used in the first segmentation result of detection, which is less than, sets
The length threshold (length threshold is the controller perturbation range being arranged according to each length of vessel of SCCT anatomy) set, if first point
The length threshold that the data group data length in result is less than setting is cut, then there may be segmentations to be broken for explanation, then second is fixed
Bit location extracts the location information being less than in the data group of length threshold, and positions the position in the first segmentation result as segmentation
Risk position.
In another embodiment, locating module 202 includes: location matches unit, for will name in the first segmentation result
Location information in information is matched with the location information in name list；Third positioning unit, if for the first segmentation knot
It names information to correspond to first location information in fruit and the location information in list is named to mismatch, then position in the first segmentation result
The position of corresponding first location information is segmentation risk position；Further include, it is abnormal to name matching unit, for the first segmentation to be tied
Information is named to be matched in fruit with the fixed name in list is named；4th positioning unit, if in the first segmentation result
It finds and names unmatched third name information with fixed, then position the position for corresponding to third name information in the first segmentation result
To divide risk position.
Wherein, location matches unit is specifically used for the position in the location information and name list in the first segmentation result
Information is matched, if existing in the location information or the first segmentation result being not present in name list in the first segmentation result
Location information has very important error in location information and name list, then there may be veins for explanation；Abnormal name matching
Unit is specifically used for fixing in the name information in the first segmentation result of comparison and name list and names, if the first segmentation result
It is middle exist in name list exist in unexistent fixed name or the first segmentation result and fixed name none matched life
Name, then there may be veins for explanation.
Coronary artery volume data model splitting equipment orients segmentation that may be present in segmentation through the above steps and omits, divides
Behind the position of fracture and vein noise, need that further it is judged and is repaired, so that second after being repaired is divided
As a result.
In one embodiment, repair module 203 includes: adjusting matching unit, for the segmentation in the first segmentation result
Risk position matches coronary artery volume data model by adjusting model segmentation confidence level, obtains the first name result；First
Unit is repaired, if meeting the first fixed name for the first name result, the first name result correspondence is inserted into the first segmentation
As a result.
Wherein, it adjusts matching unit and confidence level is adjusted to the segmentation risk position in the first segmentation result, can such as pass through tune
Low confidence is attempted to know the region again access evidence, obtains corresponding first name result and its corresponding data group, if the first life
Name result meets the first fixed name of above-mentioned omission, then the first reparation unit adds the first name result and its corresponding data group
Enter the first segmentation result.
In another embodiment, repair module 203 includes: data matching unit, for point in the first segmentation result
The data for cutting risk place-centric line extending direction are matched, and the first dividing body is obtained；Second repairs unit, if being used for first
The data length of dividing body is the data length of the second name information and the difference of length threshold, then passes through in segmentation risk position
First dividing body and the first segmentation result are attached by algorithm or adjusting confidence level.
Wherein, data matching unit knows the data segment for taking above-mentioned segmentation risk position, and reads its position data and divided
The center line extending direction data for cutting risk position are further known in the data in this direction and take the first dividing body, this first
Dividing body is in three dimensional local information, length of vessel, caliber and moves towards to match with segmentation risk position data in information.Obtain
Further the data length of the first dividing body is identified after one dividing body, if the data length is just the second name information
The difference of data length and length threshold then illustrates to know obtaining the first dividing body as required segmentation fracture location data, then the
Two reparation units, which are adjusted in segmentation risk position by algorithm or confidence level, connects the first dividing body and the first segmentation result
It connects.
In another embodiment, repair module 203 includes: vein judging unit, for the segmentation to the first segmentation result
Risk position data carries out vein judgement, obtains vein and determines result；Third repairs unit, if determining that result is for vein
It is then to remove the venous information divided at risk position in the first segmentation result；Wherein, vein determines including such as under type extremely
One of few: confidence level is adjusted, hand vein recognition and the judgement of vein tendency.
Wherein, vein judging unit carries out vein judgement to the segmentation risk position of the first segmentation result, what vein determined
Mode includes: that confidence level is adjusted, specifically by the way that confidence level is turned up can observe abnormal position and abnormal name situation whether disappear come
Judge whether it is vein；Hand vein recognition, specially to three dimensional local information, the blood vessel in abnormal name or abnormal position data group
Caliber information etc. carries out identification and judges whether it is vein；It further include the judgement of vein tendency, specially to abnormal name or exception bits
It sets the blood vessel in data group and moves towards information and carry out judging whether to meet vein trend characteristic.If the determination result is YES, then third is repaired
Multiple unit removes the venous information divided at risk position in the first segmentation result, the second segmentation result after being repaired.
The present invention is by carrying out risk supervision to coronary artery volume data model primary segmentation result, and there may be segmentation wind for positioning
The position of danger, then reparation improved model is carried out for the segmentation risk detected, it can effectively solve the problem that coronary artery volume data model point
The problems such as omission present in cutting, fracture and vein noise, and then automation coronary artery model accuracy is improved, to be cured to be clinical
It learns diagnosis and more valuable information is provided.
More than, only a specific embodiment of the invention, but scope of protection of the present invention is not limited thereto, and it is any to be familiar with
Those skilled in the art in the technical scope disclosed by the present invention, can easily think of the change or the replacement, and should all cover
Within protection scope of the present invention.Therefore, protection scope of the present invention should be subject to the protection scope in claims.
Claims (10)
1. a kind of coronary artery volume data model dividing method, which is characterized in that the described method includes:
Coronary artery volume data model is split, obtains including the first segmentation result for naming information；
Comparison name list and it is described include the first segmentation result locating segmentation risk position for naming information, the name arranges
Table is to move towards consolidating for information including at least the corresponding location information of coronary artery body blood vessel, length information, blood vessels caliber information and blood vessel
Fixed name set；
Matching reparation is carried out to oriented segmentation risk position, the second segmentation result after being repaired, after the reparation
Second segmentation result is the segmentation result repaired after segmentation omission, fracture and vein noise；
The second segmentation result after exporting the reparation.
2. being included the method according to claim 1, wherein being split to coronary artery volume data model
Name information the first segmentation result include:
During being split to coronary artery volume data model, classifies to coronary artery volume data model, obtain multiple classifications
Information；To the name information of one branch of each category information Corresponding matching in the multiple category information, to be wrapped
Include the first segmentation result of name information；
Or, extraction obtains center line during being split to coronary artery volume data model；To the center line and segmentation knot
Fruit is named automatically, obtains including the first segmentation result for naming information.
3. the method according to claim 1, wherein the comparison name list and it is described include name information
The first segmentation result locating segmentation risk position include:
Information will be named to match in first segmentation result with the fixed name in the name list；
Information is named with the first fixed name matched first if not finding in first segmentation result, positions described first
The position of the first name information is corresponded in segmentation result as segmentation risk position.
4. the method according to claim 1, wherein the comparison name list and it is described include name information
The first segmentation result locating segmentation risk position, comprising:
Detect the data length of each name information in first segmentation result；
Data length is less than the second name information of length threshold if it exists, then corresponds to described second in first segmentation result
The position for naming information is segmentation risk position.
5. the method according to claim 1, wherein the comparison name list and it is described include name information
The first segmentation result locating segmentation risk position, comprising:
By the location information progress in the location information named in first segmentation result in information and the name list
Match, if information is named not correspond to first location information and the location information named in list not in first segmentation result
Match, then positioning and corresponding to the position of the first location information in first segmentation result is segmentation risk position；
Or, information will be named to match in first segmentation result with the fixed name in the name list, if described
It is found in first segmentation result and names unmatched third name information with fixed, then positioned corresponding in first segmentation result
The position of the third name information is segmentation risk position.
6. according to the method described in claim 3, it is characterized in that, it is described to oriented segmentation risk position carry out matching repair
It is multiple, the segmentation result after being repaired, comprising:
Is divided by adjusting model to coronary artery volume data model by confidence level for segmentation risk position in first segmentation result
It is matched, obtains the first name result；
If the first name result meets the described first fixed name, described the is inserted by the first name result is corresponding
One segmentation result.
7. according to the method described in claim 4, it is characterized in that, it is described to oriented segmentation risk position carry out matching repair
It is multiple, the segmentation result after being repaired, comprising:
The data of segmentation risk place-centric line extending direction in first segmentation result are matched, obtain first point
Cut body；
If the data length of first dividing body is the data length of the second name information and the difference of the length threshold
Value is then carried out first dividing body and first segmentation result by algorithm or adjusting confidence level in segmentation risk position
Connection.
8. according to the method described in claim 5, it is characterized in that, it is described to oriented segmentation risk position carry out matching repair
It is multiple, the segmentation result after being repaired, comprising:
Vein judgement is carried out to the segmentation risk position data of first segmentation result, vein is obtained and determines result；
If the vein determine result be it is yes, the venous information at risk position will be divided described in first segmentation result
Removal；
Wherein, the vein determines to include such as at least one under type: confidence level is adjusted, hand vein recognition and the judgement of vein tendency.
9. a kind of coronary artery volume data model splitting equipment, which is characterized in that the coronary artery volume data model splitting equipment includes:
Divide module, for dividing coronary artery volume data model, obtains including the first segmentation result for naming information；
Locating module, for compare name list and it is described include name information the first segmentation result locating segmentation risk position
It sets, the name list is including at least the corresponding location information of coronary artery body blood vessel, length information, blood vessels caliber information and blood vessel
Move towards the fixed name set of information；
Repair module, for carrying out matching reparation to oriented segmentation risk position, the second segmentation result after being repaired,
The second segmentation result after the reparation is the segmentation result repaired after segmentation omission, fracture and vein noise；
Output module, for exporting the second segmentation result after the reparation.
10. coronary artery volume data model splitting equipment according to claim 9, which is characterized in that the segmentation module includes:
Classification cutting unit, for being carried out to coronary artery volume data model during being split to coronary artery volume data model
Classification, obtains multiple category informations；To the life of each one branch of category information Corresponding matching in the multiple category information
Name information, so that obtaining includes the first segmentation result for naming information；
Or, center line cutting unit, for during being split to coronary artery volume data model, extraction to obtain center line；
The center line and segmentation result are named automatically, obtain including the first segmentation result for naming information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811162108.5A CN109345546B (en) | 2018-09-30 | 2018-09-30 | A kind of coronary artery volume data model dividing method and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811162108.5A CN109345546B (en) | 2018-09-30 | 2018-09-30 | A kind of coronary artery volume data model dividing method and equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109345546A CN109345546A (en) | 2019-02-15 |
| CN109345546B true CN109345546B (en) | 2019-09-17 |
Family
ID=65308271
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811162108.5A Active CN109345546B (en) | 2018-09-30 | 2018-09-30 | A kind of coronary artery volume data model dividing method and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109345546B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109903297B (en) * | 2019-03-08 | 2020-12-29 | 数坤（北京）网络科技有限公司 | Coronary artery segmentation method and system based on classification model |
| CN109872336B (en) * | 2019-03-13 | 2021-07-09 | 数坤(北京)网络科技股份有限公司 | Blood vessel segmentation method, device and computer storage medium |
| CN109934816B (en) * | 2019-03-21 | 2021-05-11 | 数坤（北京）网络科技有限公司 | Method and device for complementing model and computer readable storage medium |
| CN111724413B (en) * | 2020-06-19 | 2021-09-28 | 上海杏脉信息科技有限公司 | Blood vessel tracking method, coronary image segmentation method, medium and electronic equipment |
| CN111862045B (en) * | 2020-07-21 | 2021-09-07 | 上海杏脉信息科技有限公司 | Method and device for generating blood vessel model |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103479381A (en) * | 2013-10-22 | 2014-01-01 | 霍云龙 | Method and equipment for accurately diagnosing coronary atherosclerosis |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| SU1177240A1 (en) * | 1984-04-12 | 1985-09-07 | Kanunnikov Sergej | Device for cleaning conveyer belt |
| CN104091346B (en) * | 2014-07-24 | 2017-02-15 | 东南大学 | Full-automatic CT image coronary artery calcification score calculating method |
| CN104952079A (en) * | 2015-06-19 | 2015-09-30 | 四川大学 | Skull segmenting method based on DICOM (Digital Imaging and Communications in Medicine) sequence |
| CN104933751B (en) * | 2015-07-20 | 2017-10-20 | 上海交通大学医学院附属瑞金医院 | The enhanced object plotting method of cardiovascular coronary artery and system based on local histogram |
| CN105389810B (en) * | 2015-10-28 | 2019-06-14 | 清华大学 | The identifying system and method for plaque within blood vessels |
| CN106127750B (en) * | 2016-06-20 | 2019-07-30 | 中国科学院深圳先进技术研究院 | A kind of CT images body surface extracting method and system |
| CN107527341B (en) * | 2017-08-30 | 2020-05-19 | 上海联影医疗科技有限公司 | Method and system for processing angiography image |
| CN108335284B (en) * | 2018-01-09 | 2022-06-28 | 北京理工大学 | Coronary vessel center line matching method and system |
-
2018
- 2018-09-30 CN CN201811162108.5A patent/CN109345546B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103479381A (en) * | 2013-10-22 | 2014-01-01 | 霍云龙 | Method and equipment for accurately diagnosing coronary atherosclerosis |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109345546A (en) | 2019-02-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109345546B (en) | A kind of coronary artery volume data model dividing method and equipment | |
| US9679389B2 (en) | Method and system for blood vessel segmentation and classification | |
| KR102050649B1 (en) | Method for extracting vascular structure in 2d x-ray angiogram, computer readable medium and apparatus for performing the method | |
| US20100296718A1 (en) | Soft tissue segmentation using a bone atlas | |
| CN111932554B (en) | Lung vessel segmentation method, equipment and storage medium | |
| CN111145206A (en) | Liver image segmentation quality evaluation method and device and computer equipment | |
| CN106097298A (en) | The coronary artery automatic segmentation divided based on spherical space and anatomic landmarks method | |
| CN109360209B (en) | Coronary vessel segmentation method and system | |
| CN103106408A (en) | Stomach computer tomography (CT) image lymph node automatic auxiliary detecting system without supervision segmentation | |
| Ren et al. | An unsupervised semi-automated pulmonary nodule segmentation method based on enhanced region growing | |
| CN109903298A (en) | Restorative procedure, system and the computer storage medium of blood vessel segmentation image fracture | |
| Tahoces et al. | Automatic detection of anatomical landmarks of the aorta in CTA images | |
| CN111553887A (en) | Method and device for obtaining evaluation information of bridge blood vessels | |
| CN107818564B (en) | Liver 3D medical image segmentation method | |
| Lin et al. | Retinal vascular tree reconstruction with anatomical realism | |
| CN109410267A (en) | A kind of coronary artery segmentation appraisal procedure and system | |
| CN107392891A (en) | Vessel tree extraction method, apparatus, equipment and storage medium | |
| CN104915961A (en) | Lump image region display method and system based on mammary X-ray image | |
| CN112733953B (en) | Lung CT image arteriovenous vessel separation method based on Non-local CNN-GCN and topological subgraph | |
| CN110766667A (en) | Damaged bone repair modeling method based on contour sample library matching and template guidance | |
| US20050256400A1 (en) | Method to identify arterial and venous vessels | |
| Shadid et al. | Bone fragment segmentation from 3D CT imagery using the Probabilistic Watershed Transform | |
| CN114445445B (en) | Artery segmentation method and device for CT image, electronic device and storage medium | |
| CN109875595B (en) | Intracranial vascular state detection method and device | |
| CN109919937B (en) | Method, system and computer storage medium for repairing blood vessel segmentation image fracture |
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 | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 100102 No. 501 No. 12, 5th floor, No. 6, Wangjing Dongyuan District 4, Chaoyang District, Beijing Patentee after: Shukun (Beijing) Network Technology Co.,Ltd. Address before: 100102 No. 501 No. 12, 5th floor, No. 6, Wangjing Dongyuan District 4, Chaoyang District, Beijing Patentee before: SHUKUN (BEIJING) NETWORK TECHNOLOGY Co.,Ltd. |
|
| CP01 | Change in the name or title of a patent holder | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 100102 No. 501 No. 12, 5th floor, No. 6, Wangjing Dongyuan District 4, Chaoyang District, Beijing Patentee after: Shukun Technology Co.,Ltd. Address before: 100102 No. 501 No. 12, 5th floor, No. 6, Wangjing Dongyuan District 4, Chaoyang District, Beijing Patentee before: Shukun (Beijing) Network Technology Co.,Ltd. |
|
| CP01 | Change in the name or title of a patent holder |