CN1765322A - Method and apparatus for embolism analysis - Google Patents
Method and apparatus for embolism analysis Download PDFInfo
- Publication number
- CN1765322A CN1765322A CN 200510099198 CN200510099198A CN1765322A CN 1765322 A CN1765322 A CN 1765322A CN 200510099198 CN200510099198 CN 200510099198 CN 200510099198 A CN200510099198 A CN 200510099198A CN 1765322 A CN1765322 A CN 1765322A
- Authority
- CN
- China
- Prior art keywords
- subtree
- organ
- embolus
- volume
- program code
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Apparatus For Radiation Diagnosis (AREA)
Abstract
Disclosed is an automated technique for analyzing the affected region due to an embolism in an organ. A segmented image of the organ vasculature is generated using image volume data received, for example, from a Computed Tomography (CT) machine. An embolus is then identified (either manually or automatically) within the segmented image, and the volume of the organ affected by the embolism is automatically determined. The volume of the organ affected by the embolism may be determined by computing a sub-tree within the segmented image, where the sub-tree comprises vessels that are distal to the identified embolus point. In one embodiment, the sub-tree is generated by determining a plane perpendicular to a vessel at the embolus point such that the sub-tree comprises a distal portion of the vasculature with respect to the plane. Unwanted overlapping trees are identified (e.g., by analyzing branch angles) and removed from the sub-tree. The volume of the organ affected by the embolism is determined by calculating a volume of the organ that is perfused by the sub-tree. The affected volume may be adjusted by scaling the volume based on the percentage occlusion of the partial embolus.
Description
The application requires the priority of the U.S. Provisional Application No.60/609521 of JIUYUE in 2004 submission on the 13rd, and this application is hereby incorporated by.
Technical field
The present invention relates generally to the embolus analysis, relate in particular to the automatic analysis in affected zone owing to thromboembolism in the organ.
Background technology
Thromboembolism is the angiemphraxis that is caused by exotic.Blood clot is the common cause of thromboembolism.The lung embolus is to be sent to the pulmonary artery (entering into the blood vessel of lung from heart) and the blood clot of occluding vascular partially or completely by blood flow.Term " embolus " is meant the tamper of occluding vascular, and thromboembolism is meant the process that embolus occurs simultaneously.
Though pulmonary infarction (PE) is the common cause of unexpected death, it normally can prevent.It is necessary using anticoagulant in time to treat saving life.But treatment contains risky equally, so correct diagnosis is only key.Computerized tomographic angiography (CTA) is accepted by people just gradually as a kind of diagnostic method, with comparing with the alternative method the pulmonary ventilation perfusion scanning such as angiopneumography, computerized tomographic angiography provides comparable or better sensitivity and specificity.CTA has the alternate diagnosis of permission to explain the advantage of patient's symptom in many cases as a kind of quick non-intrusion type method.
The image of patient behind the injection of contrast medium that obtains from 16 lamella computer tomography (CT) machines provides resolution high data, and these data allow to detect better the embolus that is arranged in sub-segment arteries.This high-resolution three-dimension data for accurately analyze pulmonary infarction to the influence of pulmonary provide may, but this evaluation may be infeasible under the situation of not using automatic operation.
In the enhanced CT image of contrast, carry out the automatic current techniques of analyzing of PE and relate to the direct detection of intra-arterial blood clot itself, or relate to the position of inferring blood clot by the visual of filling defect in the affected lung areas indirectly.In the previous case,, need carry out effective segmentation to tremulous pulse usually in order to detect the accurate position of PE.Can or finish the detection of blood clot by computer aided detection (CAD) by visualization technique then.
In another kind of CTA automatic analysis technology, directly visual in order to make filling defect, the average density of calculating and reproduction lung regional area.Showing that the lung areas be lower than average density may hint exists the upstream blood clot.The advantage of this technology is that it provides the graph-based of disease degree and seriousness.But the shortcoming of this technology is to pour in order accurately to measure, and need carry out twice scanning before and after radiography, and this needs complicated collection and twice radiation.In addition, in order to mate twice scanning, need to carry out non-rigid registration, this is quite difficult and need the expensive time.At present, after being only to carry out a radiography, the acceptable clinical practice that is used to estimate the patient that may have PE scans.
Therefore, need a kind of being used for to analyze the automatic technique of thromboembolism to the influence degree of organ according to single pass.
Summary of the invention
The invention provides a kind of being used to analyze in the organ because thromboembolism and the automatic technique in affected zone.According to one embodiment of the invention, the image volumetric data of using (for example) from computer tomography (CT) machine, to receive, the segmented image of generation organ vascular system.Then, in segmented image, discern embolus.By receiving embolus point (emboluspoint), can discern embolus (for example, here embolus point by artificial cognition) as input.Scheme can be discerned the embolus point automatically as an alternative.According to described data, determine the organ volume that influenced by thromboembolism automatically.
By calculating the subtree in the segmented image, can determine the organ volume that influenced by thromboembolism, wherein subtree comprises that those are positioned at the blood vessel of the embolus point far-end that is identified.In one embodiment, by determining that at embolus point place generating subtree with the vertical plane of blood vessel becomes, so that subtree comprises that vascular system is with respect to this planar distal portions.Because may comprising, the subtree image do not wish the overlay tree that exists, so (for example by analyzing branches angle) this overlay tree of identification and from subtree, deleting.In this, by calculating, can determine the organ volume that influenced by thromboembolism by the dabbling organ volume of subtree.In addition, because thromboembolism may only cause the partial blockage of tremulous pulse, so thereby can be by obstruction percentage convergent-divergent volume adjustment recently affected volume according to the part embolus.In addition, because may exist in patient's body, so, calculate organ by the dabbling percentage ratio of each subtree by the other embolus of identification more than an embolus, and to the summation of these percentage ratios to draw the percent of total that is subjected to the organ that all emboluses influence, can repeat this process.
With reference to the detailed description and the accompanying drawings hereinafter, these and other advantage of the present invention will be conspicuous to those skilled in the art.
Description of drawings
Fig. 1 is the high level block diagram that can carry out computer of the present invention;
Fig. 2 is the flow chart according to the performed step of embodiment of the present invention;
Fig. 3 has shown the segmentation of the lung pulse guard system S;
Fig. 4 is the flow chart of the step carried out during the tremulous pulse subtree that analysis is influenced by PE;
Fig. 5 has shown the organ vascular system with the vertical plane discerned that intersects with selected PE point;
Fig. 6 has shown tree-model;
Fig. 7 has shown the segmentation of the distal portions of selected vascular system;
Fig. 8 illustrates branches angle;
Fig. 9 illustrates and uses the branches angle analysis to determine to intersect ramose method;
Figure 10 has shown and has removed crossing tree-model and segmentation;
Figure 11 illustrates the extension of tree end branch; With
Figure 12 has shown affected organ volume.
The specific embodiment
Following description will be described the present invention according to carrying out the required treatment step of embodiment of the present invention.These steps can be carried out by the computer of suitable programming, and the structure of this computer is known in the art.For example, suitable computer can utilize known computer processor, internal storage location, storage device, computer software and other parts to realize.The high level block diagram that in Fig. 1, has shown this computer.Computer 102 comprises the processor 104 of controlling this operation by the computer program instructions of whole operations of carrying out definition computer 102.Computer program instructions can be stored in the storage device 112 (for example, disk), and can be loaded in the internal memory 110 when the needs computer program is instructed.Computer 102 also comprises one or more interfaces 106 of communicating with other device (for example local or pass through network) of being used for.Computer 102 also comprises I/O108, and this I/O representative allows user and computer 102 interactive devices (for example, display, keyboard, mouse, speaker, button etc.).Those of ordinary skill in the art is to be understood that the enforcement of actual computer also will comprise other parts, and is to be understood that Fig. 1 is the senior expression of some parts of this computer for for the purpose of illustrating.In addition, those of ordinary skill in the art is to be understood that treatment step described herein also can use specialized hardware to realize that the circuit of this specialized hardware is used to realize this treatment step by specialized designs.Scheme as an alternative, treatment step can use the various combinations of hardware and software to realize.
Fig. 2 has shown the flow chart of the step of carrying out according to an embodiment of the present invention.Originally will provide the brief general introduction of these steps.Further details about each step of Fig. 2 will be provided after ensuing general the description.At first, in step 202, receive image volume (I) data.After the reception of the image volumetric data in step 202, in step 204, carry out segmentation and analysis to the lung pulse guard system.Then, identification is positioned at the intrasystem PE of segmented pulmonary vessel position in step 206.This identification at PE position can input manually after the inspection of the image of segmented pulmonary vessel system, and perhaps the PE position can be automatically recognized.In step 208, analyze the lung volume of vascular subtree that influenced by PE to determine influenced by PE.
To use following symbol that the further details of step shown in Figure 2 is described in more detail now.The original image volume is represented with I.(x, y z) provide, wherein (x, y, z) 3 dimension coordinates at expression PE position by p=at selected PE position.The volume that comprises segmentation vascular system in the lung is represented with S.At last, segmentation subtree volume is represented with S '.
In step 204, carry out segmentation and feature description to the lung pulse guard system.In one embodiment, as Pichon E, Novak CL, Kiraly AP, Naidich DP carries out described step in such described in SPIE medical imaging proceedings in 2004 the 5367th the 161st to the 170 page of article of delivering of volume " A novelmethod for pulmonary emboli visualization from high-resolution CT images ", and the document is incorporated herein by reference.The following work of this technology.At first, create lung mask (mask).In trachea, select seed points at first.Then, in the growth of this seed points place execution area, up to whole lung by segmentation.In order to be full of lung by air flue, this region growing comprises high threshold.Then, segmented image is carried out expansion and corrosion, to fill the vacuum that causes by fluid filled zone (such as blood vessel).In order to prevent that near other structure rib and the thoracic wall is included in the mask, erosion operator should be slightly larger than the expansion operator.Then, by comprising that all carry out segmentation greater than the voxel of threshold value to the lung blood vessel in the lung mask.Because the existence of PE will hinder contrast agent flow to some blood vessel, so threshold value selected the blood vessel that has contrast agent and do not have contrast agent to comprise.Then, segmental structure is carried out the coupling part labelling.Structure with small size is left out.The result is the segmentation of the lung pulse guard system S.The limitation of this processing is and possible also comprises pulmonary vein or other compact texture together with tremulous pulse.Resulting segmentation S is shown as 300 in Fig. 3.Segmentation 300 has shown the profile of lung 302 and vascular system 304.
Any other known method also can be used to produce segmentation.For example, also can use the line filter described in medical image analysis in 1998 the 2nd volume the 2nd phase the 143rd to 168 page of article of delivering " Three-dimensional multi-scale line filter forsegmentation and visualization of curvilinear structures inmedical images " as people such as Sato Y.In addition, the tree analysis to line filter output can be used to improve accuracy.
After generating segmentation, calculate signed distance map Ds at the lung blood vessel.Ds has provided each voxel in the S to the distance of nearest surface point.Because bigger tremulous pulse has bigger radius, so bigger tremulous pulse will have bigger Ds in their center.As described in greater detail below such, this information is used to calculate subtree.
The identification of the intrasystem PE of step 206 (Fig. 2) expression segmented pulmonary vessel position p.This PE position is shown as a little 306 in Fig. 3.This point can be discerned and indication (for example by the internist) artificially, perhaps can be the output of automatic detection algorithm.In either case, step 206 represents that all PE position p is the input of treatment step.For example, the input at PE position can be by input-output apparatus 108 (if artificial cognition) or by interface 106 (if identification automatically).
In processing procedure, can utilize following data in this:
A) the enhanced CT image I of primary contrast;
B) comprise the segmentation of the lung blood vessel S of range mark Ds; With
C) be used for the further some p of analysis in the segmented image.
In this, processing can be proceeded being subjected to the analysis of the tremulous pulse subtree that PE influences with step 208.In step 208, analyze the subtree influenced by PE, so that determine the lung volume that influenced by thromboembolism.In conjunction with the flow chart among Fig. 4, further describe processing according to step 208.As shown in Figure 4, the subtree analytical procedure is made up of 4 steps (402-408).At first, in step 402, determine and the vertical bisecting plane of vessel directions at a p place.This plane is used to carry out restricted region growing in step 404, is called as the distal portions of the tree of subtree with separation.In step 406, calculate tree-model by method according to subtree, and tree-model is analyzed to eliminate crossing blood vessel based on skeletonizing.In step 408, determine affected lung volume.The further details of each step among Fig. 4 will be described below.
Step 420 determine with blood vessel on the crossing vertical plane of selected PE point.This plane is represented as plane 502 in Fig. 5, the PE point is represented as 504.According to described step, under the situation of the given reconnaissance p of institute 504, near a p, set up the fixed size sub-volumes of segmentation S.Should be noted that on near-end or distal direction described fixed size sub-volumes only comprises local tree structure rather than whole tree.Then, by known tree computational methods, the segmentation vascular system in the described sub-volumes is carried out modeling based on skeletonizing.People such as A.P.Kiraly have described described tree computational methods based on skeletonizing at the paper " Three-Dimension Path Planning for Virtual Bronchoscopy " that in JIUYUE, 2004 is delivered in the 1365th to 1379 page of the 23rd the 9th phase of volume of IEEE medical imaging journal, this paper is incorporated herein by reference.This method is in this following general description of carrying out.
The tree computational methods are calculated the tree-model that provides vascular system segmented image and root position.In Fig. 6, shown simple tree-model.Tree is made up of many continuous branches.Each branch is made up of many positions again.The root of root position definition tree, and determine all ramose filiations.Branch part is the position that occurs bifurcated between two branches.Distal portion finds at the ramose end points place that does not have sub-branch.At last, all other positions all are called as the observation position.
Following step is used for determining tree-model.Under the situation that provides a segmental structure, at first calculate its 3D skeleton.This operation converts segmentation to the structure of the plain thickness of being made up of the 3D line of bifurcated of monomer.Then, skeleton is stored in the tree-model form, in this tree-model form, find branch and branch point.The root of tree is determined in the root position.Each ramose a plurality of position is to constitute ramose voxel location.This initial tree-model comprises the false ramification that causes owing to the discreteness of data and sectional roughening probably.For improved model, use and delete false ramification based on the standard of size.The position, position also can be improved to the plain level of daughter.At last, be that the basis serves as that specify and the vertical direction of branch direction at ramose each position of residue with the adjacent portion bit position.
Described method is applied near the sub-volumes that is obtained the p.We are interested to be the vertical plane that obtains at position p place.To elect the root position as apart from p point farthest in the segmentation simply.Should be noted that this root position may be incorrect with respect to far-end and the proximal part of tree.Real root position should be positioned on the closest branch of tree.But the root bit position does not influence the vertical plane at position p place that is calculated.Under the situation of the model that calculates providing, get apart from the nearest position of p and be considered as the observed direction at described position vertical with this plane.
Next procedure (404 among Fig. 4) is used for defined plane to cut subtree from the remainder of tree.Step 404 is following carries out.Suppose plane 502 bisection blood vessels, still have such problem: which side of tree comprises interested subtree, and which side comprises the nearest blood vessel towards heart.The purpose of step 404 is the segmentations that generate the distal portions of selected vascular system, and described segmentation is called as S '.The segmental structure S that is positioned at p place, position is carried out two kinds of rule-based region growing operations, carry out a kind of rule-based region growing operation in planar each side.Carry out confined in two ways standard 3D region growing operation.At first, it must limit by the segmentation that defines in S.Secondly, the plane 502 of previous definition can not be passed in the zone.Less one is counted as the far-end tree in two zones.Final result is to determine far-end subtree S ' S, and this is the basis of further handling.This far-end subtree S ' is shown as 700 in Fig. 7.It should be noted that this seed tree can comprise the additional blood vessel 704 that intersects owing to seeming of causing of partial volume effect.These additional blood vessels that intersect are handled in next procedure.
In step 406, calculate tree-model (as mentioned above) by method according to subtree, and analyze this tree-model to eliminate crossing blood vessel based on skeletonizing.Suppose subtree S ' 700, its tree construction is determined by top integrating step 402 described tree computational methods based on skeletonizing.This tree based on skeletonizing is shown as in Fig. 7 and is positioned at endovascular line (for example, 706).As mentioned above, step 402 is only calculated near the tree construction the reconnaissance p of institute, rather than whole subtree.The target of step 406 provides a seed tree model.The branch end points nearest apart from the reconnaissance p of institute is elected to be the root position of model automatically.But tree computational methods hypothesis constitutes real vascular tree as the segmentation tubular structure that input provides, and does not comprise any overlay structure.This is not a situation of including all given subtrees, because the lung blood vessel may be owing to partial volume effect comprises significantly overlapping.Therefore, as long as this tree is intersected in S ', the model that is calculated will be brought incorrect branch in the contiguous vascular tree into.
Given sectional tree-model can determine that the blood vessel that is caused by adjacent blood vessel intersects.At first, as shown in Figure 8, determine each ramose branches angle 802.Fig. 9 illustrates and is used to discern and eliminate the method that intersects blood vessel.Distal arterial tree 902 has usually and parent branch forms bifurcation structure greater than 90 branches angles of spending.For example, sub-branch 904 has the branches angle 906 greater than 90 degree.Consider the blood vessel 908 of subtree 910 now, this blood vessel is interior crossing at S ', therefore is trapped in the tree construction.This branches angle 912 that will produce less than 90 degree that intersects, generation simultaneously have the branch born of the same parents 914 of complementary branches angle 916 (the error surplus allows to regard these two branches angles as complementary).Therefore, according to step 406, by 1) sub-branch has branches angle for acute angle with respect to parent branch; With 2) branch born of the same parents has complementary branches angle, and blood vessel is intersected in identification.Therefore, according to described test, branch 908 and 914 is identified as crossing blood vessel.The crossing branch that is identified is deleted from model.Should be noted that from tree-model that deletion branch also needs to delete as its sub-branch is captured to all branches in the model.Therefore, in the example of Fig. 9, parent branch 908 and branch 918 are also with deleted, because with respect to the root position, tree construction is modeled as the sub-branch of branch 908 with them.Identification and deletion continue to carry out in the mode of iteration, and be deleted from model up to the whole adjunct tree structure that intersects blood vessel.
In Figure 10, be shown as 1002 without any the final tree-model and the segmentation of intersecting, and can be used to improve blood vessel subtree in the middle definition of S '.Each ramose each position links together by Ds and distance value.As mentioned above, distance value has provided the shortest path value from given segmentation voxel to blood vessel surface.Therefore, this value is the radius of the maximum spheroid that can comprise in special voxel location punishment section.The availability of these distance values allows tree-model to rebulid S ' by placing sizeable spheroid at each position of tree-model and catching S with the part that these spheroids intersect.Because it is deleted from model to intersect tree, so they can be in the middle appearance of this S ' that rebulids.Then, final S ' only comprises the blood vessel subtree that intersects of not having as shown in figure 10.Therefore, the initial of S ' determines it is not final, and further improved by obtainable high level information from tree-model.
In this, step 408 can continue to quantize the lung volume that influenced by PE.408 pairs of this quantization steps are estimated by the dabbling lung volume of subtree zone.In order to estimate this zone more accurately, need comprise near the detailed subtree of the blood vessel that thoracic wall is.Although high-resolution ct can extract little blood vessel, in most of the cases segmentation subtree can not reach thoracic wall.But segmentation is fully near thoracic wall, so that well being similar to affected volume to be provided.
As described in Figure 11, each end branch in the tree-model (just, not having the branch of sub-branch) (1102,1104) is extended linearly, runs into the lung border up to it at thoracic wall 1106 places.As mentioned above, each position all has associated observed direction, and this observed direction defines the direction of blood vessel at this point.Determined bearing of trend in the observed direction that each distal portion (1108,1110) of each end branch is located.
In order to determine affected lung volume, calculate the 3 dimension convex closures 1112 that extend tree according to known manner.This convex closure 1112 has defined affected lung zone.Then, (by the number of the voxel in the convex closure is counted) measures this regional volume, and with this volume divided by the volume (calculating) of whole lung by the number of the voxel in the lung mask is counted to obtain the percentage ratio of affected lung.Because branch is near thoracic wall, so aforesaid ramose extension is a kind of acceptable estimation.Any other branch does not intersect in the outside of convex closure probably.
Can pass through I/O 108 (Fig. 1) (for example, computer display monitor) by the image that described technology produced and be shown to the user, so that the graphical feedback of segmentation and quantized result to be provided.In advantageous embodiment, as shown in figure 12, segment arteries 1202 can show with transparent color, and the convex closure 1204 of sub-segmentation tree is used to make affected area visualization.In addition, the transparent view of lung 1206 provides the visual display of affected lung volume.
For the sake of clarity, in conjunction with single PE position and the subtree of being extracted described embodiment is described.But some patients have a plurality of emboluses.PE position for selecting in addition can repeat described step, wherein extracts new subtree at each position.Can be sued for peace together to indicate affected total lung volume by the volume that each subtree faced toward.Be located immediately at indicated PE under the situation in downstream of another PE, a subtree will be completely contained in another subtree.This situation can detect from tree is calculated.In this case, when calculating affected overall area, the volume with less tree is not added on the bigger tree.
Should be understood that in some cases, the lung embolus is obstructing arterial partly only.In this case, blood still can flow through blood clot, thereby allows affected subregion partly to be poured into by blood.In these cases, can calculating lung, to be subjected to the embolus effect be exceedingly useful.The influence of the embolus that following embodiment estimating part ground blocks.Under the situation of given PE position and segmentation blood vessel, from tremulous pulse, be partitioned into blood clot, so that calculate the cross-sectional area Ac of blood clot perpendicular to the segmentation blood vessel.Also calculate the cross-sectional area Av of the blood vessel that blood clot occurs according to blood vessel segmentation.As the maximum of Ac/Av, calculate and block percentage ratio.Influence degree to lung is by blocking the affected subregion of percentage ratio convergent-divergent.In the example shown in Figure 12, be 80% if block percentage ratio, the influence degree to lung is 4.6% so, or 5.7% 80%.
Top detailed description is appreciated that it all is being illustrative and exemplary aspect each, but not restrictive, and scope of invention disclosed herein should not determined according to this detailed description, but determine according to the claim of explaining as the four corner of permitting according to Patent Law.Should be understood that embodiment shown and that describe has only illustrated principle of the present invention here, and those skilled in the art can realize various modifications under the situation that does not deviate from the spirit and scope of the present invention.Under the situation that does not deviate from the spirit and scope of the present invention, those skilled in the art can realize various further feature combinations.For example, use here pulmonary infarction as an illustration the property embodiment invention has been described.But the present invention is not limited to pulmonary infarction, but can be applicable to the thromboembolism of any kind.In addition, CT data property embodiment is as an illustration used in described description.But, the present invention can be applicable to any kind 3 the dimension medical images, such as magnetic resonance.
Claims (40)
1. one kind is used to analyze in the organ because thromboembolism and the automated process in affected zone may further comprise the steps:
Utilize image volumetric data to generate the segmented image of organ vascular system;
Discern the embolus point in the described segmented image; With
Automatically determine the organ volume that influenced by described thromboembolism.
2. method according to claim 1, wherein said organ is a lung.
3. method according to claim 1, the step of wherein said identification embolus point comprise imports reception with described embolus point as the user.
4. method according to claim 1, the step of wherein said identification embolus point comprise the described embolus point of automatic identification.
5. method according to claim 1, wherein said image volumetric data comprises the computer tomography data.
6. method according to claim 1, wherein saidly determine to be subjected to the step of the organ volume that described thromboembolism influences further to may further comprise the steps automatically:
According to the obstruction percentage ratio of part embolus, the organ volume that convergent-divergent is influenced by described thromboembolism.
7. method according to claim 1, wherein saidly determine to be subjected to the step of the organ volume that described thromboembolism influences further to may further comprise the steps automatically:
Calculate the subtree in the described segmented image.
8. the step of the subtree in the method according to claim 7, the described segmented image of wherein said calculating further may further comprise the steps:
Identification is positioned at the blood vessel of described embolus point far-end.
9. the step of the subtree in the method according to claim 7, the described segmented image of wherein said calculating further may further comprise the steps:
Determine on described embolus point place and the vertical plane of blood vessel;
Wherein said subtree comprises that described vascular system is with respect to described planar distal portions.
10. method according to claim 7, wherein said subtree image comprises a plurality of overlay trees that cause owing to the adjacent blood vessel in the described image, described method further may further comprise the steps:
Discern described overlay tree; With
The described overlay tree of deletion from described subtree.
11. method according to claim 10, the step of wherein said identification overlay tree further may further comprise the steps:
Analyze the branches angle in the described subtree image.
12. method according to claim 11, the step of wherein said identification overlay tree further may further comprise the steps:
If sub-branch has for the branches angle of acute angle and branch born of the same parents have complementary branches angle, then discern overlapping blood vessel.
13. method according to claim 7 wherein saidly determines to be subjected to the step of the organ volume that described thromboembolism influences further to may further comprise the steps:
Calculate volume by the dabbling described organ of described subtree.
14. method according to claim 13, wherein said calculating further may further comprise the steps by the step of the volume of the dabbling described organ of described subtree:
The edge that extends to described organ by the end branch with described subtree produces the extension subtree; With
Calculate the convex closure of described extension subtree.
15. be used to analyze in the organ because thromboembolism and the equipment in affected zone comprises:
Be used to utilize image volumetric data to produce the device of the segmented image of organ vascular system;
Be used to discern the device of the embolus point in the described segmented image; With
Be used for determining automatically being subjected to the device of the organ volume that described thromboembolism influences.
16. equipment according to claim 15, the wherein said device that is used to discern embolus point comprises the device that is used for described embolus point is imported as the user reception.
17. equipment according to claim 15, the wherein said device that is used to discern embolus point comprises the device that is used for discerning automatically described embolus point.
18. equipment according to claim 15, wherein said image volumetric data comprises the computer tomography data.
19. equipment according to claim 15 wherein saidly is used for determining automatically being subjected to the device of the organ volume that described thromboembolism influences further to comprise:
Be used for according to the obstruction percentage of part embolus recently convergent-divergent be subjected to the device of the organ volume that described thromboembolism influences.
20. equipment according to claim 15 wherein saidly is used for determining automatically being subjected to the device of the organ volume that described thromboembolism influences further to comprise:
Be used to calculate the device of the subtree in the described segmented image.
21. equipment according to claim 20, the wherein said device that is used to calculate the subtree in the described segmented image further comprises:
Be used to discern the device of the blood vessel that is positioned at described embolus point far-end.
22. equipment according to claim 20, the wherein said device that is used to calculate the subtree in the described segmented image further comprises:
Be used to determine at described embolus point place and the vertical planar device of blood vessel;
Wherein said subtree comprises that described vascular system is with respect to described planar distal portions.
23. equipment according to claim 20, wherein said subtree image comprises a plurality of overlay trees that cause owing to the adjacent blood vessel in the described image, and described device further comprises:
Be used to discern the device of described overlay tree; With
Be used for from the device of the described overlay tree of described subtree deletion.
24. equipment according to claim 23, the wherein said device that is used to discern overlay tree further comprises:
Be used to analyze the device of the branches angle in the described subtree image.
25. equipment according to claim 24, the wherein said device that is used to discern overlay tree further comprises:
Be used for having the device of identification overlay tree when having complementary branches angle for the branches angle of acute angle and branch born of the same parents in sub-branch.
26. equipment according to claim 20 wherein saidly is used to determine to be subjected to the device of the organ volume that described thromboembolism influences further to comprise:
Be used to calculate device by the volume of the dabbling described organ of described subtree.
27. equipment according to claim 26, the wherein said device that is used to calculate the volume by the dabbling described organ of described subtree further comprises:
Be used for generating the device that extends subtree by the edge that the end branch with described subtree extends to described organ; With
Be used to calculate the device of the convex closure of described extension subtree.
28. a computer-readable medium comprises that being used to of being stored analyzed in the organ because thromboembolism and the computer program code in affected zone, when being carried out by processor, described computer program code defines following steps:
Utilize image volumetric data to produce the segmented image of organ vascular system;
Discern the embolus point in the described segmented image; With
Automatically determine the organ volume that influenced by described thromboembolism.
29. computer-readable medium according to claim 28, the computer program code of the step of wherein said definition identification embolus point comprise that definition imports described embolus point the computer program code of the step of reception as the user.
30. computer-readable medium according to claim 28, the computer program code of the step of wherein said definition identification embolus point comprise the definition computer program code of the step of the described embolus point of identification automatically.
31. computer-readable medium according to claim 28, wherein said image volumetric data comprises the computer tomography data.
32. computer-readable medium according to claim 28, wherein said definition determine to be subjected to the computer program code of the step of the organ volume that described thromboembolism influences further to comprise the computer program code that defines following steps automatically:
According to the obstruction percentage ratio of part embolus, the organ volume that convergent-divergent is influenced by described thromboembolism.
33. computer-readable medium according to claim 28, wherein said definition determine to be subjected to the computer program code of the step of the organ volume that described thromboembolism influences further to comprise the computer program code that defines following steps automatically:
Calculate the subtree in the described segmented image.
34. calculating the computer program code of the step of the subtree in the described segmented image, computer-readable medium according to claim 33, wherein said definition further comprise the computer program code that defines following steps:
Identification is positioned at the blood vessel of described embolus point far-end.
35. calculating the computer program code of the step of the subtree in the described segmented image, computer-readable medium according to claim 33, wherein said definition further comprise the computer program code that defines following steps:
Determine on described embolus point place and the vertical plane of blood vessel;
Wherein said subtree comprises that described vascular system is with respect to described planar distal portions.
36. computer-readable medium according to claim 33, wherein said subtree image comprises a plurality of overlay trees that cause owing to the adjacent blood vessel in the described image, and described computer-readable medium further comprises the computer program code that defines following steps:
Discern described overlay tree; With
The described overlay tree of deletion from described subtree.
37. computer-readable medium according to claim 36, the computer program code of the step of wherein said definition identification overlay tree further comprises the computer program code that defines following steps:
Analyze the branches angle in the described subtree image.
38. according to the described computer-readable medium of claim 37, the computer program code of the step of wherein said definition identification overlay tree further comprises the computer program code that defines following steps:
If sub-branch has for the branches angle of acute angle and branch born of the same parents have complementary branches angle, then discern overlay tree.
39. computer-readable medium according to claim 33, wherein said definition determine to be subjected to the computer program code of the step of the organ volume that described thromboembolism influences further to comprise the computer program code that defines following steps:
Calculate volume by the dabbling described organ of described subtree.
40. according to the described computer-readable medium of claim 39, the computer program code of step that the volume of described organ is calculated in wherein said definition further comprises the computer program code that defines following steps:
The edge that extends to described organ by the end branch with described subtree generates the extension subtree; With
Calculate the convex closure of described extension subtree.
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US60952104P | 2004-09-13 | 2004-09-13 | |
| US60/609521 | 2004-09-13 | ||
| US11/202801 | 2005-08-12 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN1765322A true CN1765322A (en) | 2006-05-03 |
Family
ID=36741586
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200510099198 Pending CN1765322A (en) | 2004-09-13 | 2005-09-13 | Method and apparatus for embolism analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1765322A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102208105A (en) * | 2010-03-31 | 2011-10-05 | 富士胶片株式会社 | Medical image processing technology |
| CN101541242B (en) * | 2006-11-30 | 2012-06-13 | 皇家飞利浦电子股份有限公司 | Visualizing a vascular structure |
| CN101548296B (en) * | 2006-06-16 | 2013-04-24 | 皇家飞利浦电子股份有限公司 | Automated hierarchical splitting of anatomical trees |
-
2005
- 2005-09-13 CN CN 200510099198 patent/CN1765322A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101548296B (en) * | 2006-06-16 | 2013-04-24 | 皇家飞利浦电子股份有限公司 | Automated hierarchical splitting of anatomical trees |
| CN101541242B (en) * | 2006-11-30 | 2012-06-13 | 皇家飞利浦电子股份有限公司 | Visualizing a vascular structure |
| CN102208105A (en) * | 2010-03-31 | 2011-10-05 | 富士胶片株式会社 | Medical image processing technology |
| CN102208105B (en) * | 2010-03-31 | 2017-03-01 | 富士胶片株式会社 | Medical Image Processing |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7583829B2 (en) | Method and apparatus for embolism analysis | |
| US9471989B2 (en) | Vascular anatomy modeling derived from 3-dimensional medical image processing | |
| JP6877868B2 (en) | Image processing equipment, image processing method and image processing program | |
| US8244012B2 (en) | Computer aided detection of pulmonary embolism with local characteristic features in CT angiography | |
| US7272250B2 (en) | Vessel segmentation with nodule detection | |
| US8913060B2 (en) | Systems and methods for extracting a curve-skeleton from a volumetric image of a vessel | |
| US7961967B2 (en) | Adaptive density mapping in computed tomographic images | |
| US20050110791A1 (en) | Systems and methods for segmenting and displaying tubular vessels in volumetric imaging data | |
| US7787673B2 (en) | Method and apparatus for airway detection and segmentation using 3D morphological operators | |
| US8165376B2 (en) | System and method for automatic detection of rib metastasis in computed tomography volume | |
| CN111528871A (en) | Method and system for whole body bone removal and vessel visualization in medical images | |
| WO2008097520A2 (en) | System and method for computer aided detection of pulmonary embolism in tobogganing in ct angiography | |
| CN1853569A (en) | Vessel boundary deteching method and device | |
| US7486812B2 (en) | Shape estimates and temporal registration of lesions and nodules | |
| US7480401B2 (en) | Method for local surface smoothing with application to chest wall nodule segmentation in lung CT data | |
| US8224057B2 (en) | Method and system for nodule feature extraction using background contextual information in chest x-ray images | |
| Ng et al. | Medical image segmentation using watershed segmentation with texture-based region merging | |
| CN114972376A (en) | Coronary calcified plaque segmentation method, segmentation model training method and related device | |
| CN114943699B (en) | Segmentation model training method, coronary calcified plaque segmentation method and related device | |
| WO2022164374A1 (en) | Automated measurement of morphometric and geometric parameters of large vessels in computed tomography pulmonary angiography | |
| CN1765322A (en) | Method and apparatus for embolism analysis | |
| CN111429457B (en) | Intelligent evaluation method, device, equipment and medium for brightness of local area of image | |
| Skibbe et al. | Efficient Monte Carlo image analysis for the location of vascular entity | |
| KR101474162B1 (en) | Automatic Segmentation system and the method of Ground-glass Opacity Nodule in Chest CT Images | |
| Wang et al. | Airway segmentation for low-contrast CT images from combined PET/CT scanners based on airway modelling and seed prediction |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C41 | Transfer of patent application or patent right or utility model | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20061222 Address after: American Pennsylvania Applicant after: American Siemens Medical Solutions Inc. Co-applicant after: Siemens AG Address before: new jersey Applicant before: Siemens Corporate Research, Inc. Co-applicant before: Siemens AG |
|
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |