CN1879128A - Dynamic crop box determination for optimized display of a tube-like structure in endoscopic view - Google Patents
Dynamic crop box determination for optimized display of a tube-like structure in endoscopic view Download PDFInfo
- Publication number
- CN1879128A CN1879128A CNA2004800328701A CN200480032870A CN1879128A CN 1879128 A CN1879128 A CN 1879128A CN A2004800328701 A CNA2004800328701 A CN A2004800328701A CN 200480032870 A CN200480032870 A CN 200480032870A CN 1879128 A CN1879128 A CN 1879128A
- Authority
- CN
- China
- Prior art keywords
- ray
- observation point
- data set
- area
- current observation
- 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
- Image Generation (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
Methods and systems for dynamically determining a crop box to optimize the display of a subset of a 3D data set, such as, for example, an endoscopic view of a tube-like structure, are presented. In exemplary embodiments of the present invention, a 'ray shooting' technique can be used to dynamically determine the size and location of a crop box. In such embodiments, shot rays are distributed evenly into a given volume and their intersection with the inner lumen determines the crop box boundaries. In alternate exemplary embodiments, rays need not be shot into fixed directions, but rather may be shot using a random offset which changes form frame to frame in order to more thoroughly cover a display area. In other exemplary embodiments, in order to get even better results, more rays can be shot at areas of possible error, such as, for example, where the centerline of a tube-like structure is leading to. In such embodiemnts rays need not be distributed evenly, but can be varied in space and time, i.e. In each frame the program can, for example, shoot out a different number of rays, in different directions, and the distribution of those rays could be in different pattern. Because, in exemplary embodiemnts, a dynamically optimized crop box encloses only the portion of the 3D data set which is actually displayed, processing cycles and memory usage are minimized.
Description
The cross reference of other applications
The application requires to enjoy the rights and interests of following U.S. Provisional Patent Application, wherein each disclosed content whole is incorporated into herein as a reference: the sequence number 60/517 that each is all submitted on November 3rd, 2003, the sequence number 60/562,100 that on April 14th, 043 and 60/516,998 and 2004 submitted to.
Technical field
The present invention relates to the field of the mutual demonstration of 3D data set, and, more specifically, relate to and dynamically determine the demonstration of crop box (crop box) with tubular structure among the optimization endoscope figure.
Background technology
Health care expert and researchist are usually interested in the inside of observing tubular anatomical structure, as for example blood vessel of person under inspection's health (for example main artery) or digestive system inner-cavity structure (for example colon) and so on.In history, the user can to observe the method that only has of these structures be by inserting endoscope probe and camera, as in traditional colonoscopy or the endoscopy.Appearance along with complicated imaging technique, picture for example magnetic resonance imaging (" MRI "), echo-planar imaging (" EPI "), computed tomography (" CT ") and newer electrical impedance tomography scan (" EIT ") and so on, can obtain a plurality of images of various luminal organ and construct 3D body (3D volume) thus.Draw (render) these bodies can for then radiologist or other diagnosticians, be used for non-intrusive inspection is done in the inside of patient pipe.
For example, in colonoscopy, can come editosome data (volumetric data) collection by one group of underbelly CT section (usually in the scope that 300-600 opens, but also can be) more than 1000.These CT sections can for example be expanded by various interpolation methods, to produce the said three-dimensional body of using the conventional bulk rendering technique to draw.Use such technology, can on suitable display, show 3-D data set, and make the user can carry out the fictitious tour of patient's colon, so just exempted the needs that insert endoscope.Such process is called as " virtual coloscope inspection ", and has become recently and can use the patient.
Although the obvious advantage of its Noninvasive is arranged, still there is some intrinsic in virtual coloscope inspection inconvenience and difficult.In general, these problems all exist in the virtual examination of any tubular anatomical structure that uses conventional art.
For example, traditional virtual coloscope inspection places colonic lumen inside to user's observation point, and moves this observation point along computing center's line in whole inside usually.In such demonstration, suppose it is that the standard monoscopic shows, then lack depth cue (depth cue) usually.Like this, the key property of colon may be not in sight, and may still not discover problem area.
In addition, the typical case of tubular anatomical structure demonstration only illustrates part-structure among the endoscope figure on display screen.Usually, endoscope figure is the sub-fraction of corresponding whole tubular structure only, for example press sweep volume 2% to 10% and press tubular structure length 5% to 10% or more.In the middle of the demonstration of such endoscope figure, if display system is drawn whole colon and only shown its sub-fraction, then such technology is consuming time but also poor efficiency not only.If system can determine and only draw and will then just can be saved a large amount of processing times and storage space like this by the part of actual displayed for user or observer.
Further, as known in the Volume Rendering Techniques, need the voxel (voxel) of drafting and demonstration many more, the demand of computational resource aspect is just high more.The demand of computational resource aspect is also proportional with given user-selected sharpness level, for example amplifies or pass through to improve rendering quality by increasing digital picture.If select higher sharpness, then during sampler body, must produce the polygon of greater number.In the time will sampling more polygon, just need to draw more pixel (and generally speaking each pixel on the screen can be repeated to fill many times), and filling rate can reduce.Under high-caliber sharpness, the render speed of the body segment that a large amount of like this input data can slow down observed, and for example the user is after observation point is moved in new position, the image of demonstrations to be filled such as needs.
On the other hand, wishing usually has higher sharpness, and in fact usually needs, so that can carry out careful diagnosis or analysis by assisted user.In addition, if the desired depth prompting, for example by draw the body of being paid close attention to three-dimensionally, the polygonal number of sampling that then needs to be input to rendering algorithm can double, and required storer also doubles when drawing thereby do.
In general, observe big 3D data set when the user is once mutual partially, wherein the part of observing at any one time is the sub-fraction of whole data set, but described part can not a priori be determined, to all such situations, above-mentioned prior art problems all exists.As handle the not voxel of actual displayed, from just showing the processing of voxel and draw remove needed computational resource, except other difficulties are also introduced waiting status etc., unless adopt remedying of certain mode, otherwise so mutual observation just there is not use.
Like this, needed in this technical field is exactly the process optimizing that shows big 3D data set, and wherein, in many given moment, just checked body portion only is the subclass of complete body.Such optimization should effectively utilize computational resource more, and needs under a large amount of high resolving power of calculating at each voxel to be drawn, and provides depth cueing, more high definition and instrument and the functional seamless no wait state that freely uses checks.
Summary of the invention
Provide and be used for dynamically determining the method and system of crop box with the demonstration of the subclass of the 3D data set of virtual endoscope figure of optimizing tubular structure for example and so on.In exemplary embodiment of the present invention, " ray emission " technology can be used in size and the position of dynamically determining crop box.In such embodiments, for example, can be in given body ray emission, and for example the intersection of they and inner chamber can be determined crop box boundaries.In exemplary embodiment of the present invention, ray does not need to be transmitted into fixing direction, but can for example use random offset to launch, and described random offset changes with the change of frame, so that can cover the viewing area more completely.In other exemplary embodiments, can launch more ray in the zone of possible error, for example the direction of the Far Range along the tubular structure center line or close direction are launched from current observation point.In such exemplary embodiment, ray can change by room and time, and wherein, for example in each frame, exemplary process for example can be launched the ray of varying number on different directions, and the distribution of those rays can be in different patterns.Because the crop box of dynamic optimization only surrounds the actual displayed part that the 3D data set is put at any time, so employed processing cycle and memory usage can significantly be minimized during the drawing data collection.
From following accompanying drawing and various detailed description of illustrative embodiments, further feature of the present invention, its character and various advantage will be more clear.
Other target of the present invention and advantage will partly be set forth in explanation subsequently, and can partly become clear by instructions, perhaps can be by enforcement of the present invention is known.By means of key element that particularly points out in the accessory claim and combination, can realize and obtain target of the present invention and advantage.
It being understood that aforementioned general remark and following detailed description all are exemplary with indicative, and be not as claims, to limit the present invention.
Description of drawings
Fig. 1 shows part human colon's example virtual endoscope figure;
Fig. 1 (a) is the grayscale version of Fig. 1;
Fig. 2 shows the exemplary current viewing mask as the sub-fraction demonstration of exemplary human colon's total figure;
Fig. 2 (a) is the grayscale version of Fig. 2;
Fig. 3 has described according to the exemplary ray among the current virtual endoscope figure of being transmitted into of exemplary embodiment of the present invention;
Fig. 3 (a) is the grayscale version of Fig. 3;
Fig. 4 has described the side view of the divergent-ray of Fig. 3;
Fig. 4 (a) is the grayscale version of Fig. 4;
Fig. 5 has shown the exemplary crop box that is limited according to exemplary embodiment of the present invention, in order to surround all points of impact of divergent-ray;
Fig. 6 has described the one group of exemplary equally distributed ray point of impact that is used to limit crop box according to exemplary embodiment of the present invention, does not wherein draw out the part farthest of colon;
Fig. 6 (a) is the grayscale version of Fig. 6;
Fig. 7 shows the one group of exemplary hit point that expands Fig. 6 according to the utilization of exemplary embodiment of the present invention about the equally distributed one group of additional point of impact in the end of shown center line;
Fig. 7 (a) is the grayscale version of Fig. 7;
Fig. 8 (a)-(d) shows the generation of the crop box and the crop box that the cone (viewing frustrum) is aimed at of axon aligning according to various embodiments of the present invention;
Fig. 9 (a) and (b) show the exemplary big sampled distance that is used to draw body (with the polygon of little respective numbers);
Fig. 9 (c) and (d) be respectively Fig. 9 (a) and grayscale version (b);
Figure 10 (a) and (b) shown the less sampled distance that is used to draw body (with the polygon of bigger respective numbers) with respect to Fig. 9;
Figure 10 (c) and (d) be respectively Figure 10 (a) and grayscale version (b);
Figure 11 (a) and (b) shown the sampled distance littler that be used to draw body (with the polygon of bigger respective numbers) with respect to Figure 10;
Figure 11 (c) and (d) be respectively Figure 11 (a) and grayscale version (b);
Figure 12 (a) and (b) show the sampled distance littler that be used to draw body (with the polygon of bigger respective numbers) with respect to Figure 11;
Figure 12 (c) and (d) be respectively Figure 12 (a) and grayscale version (b);
Figure 13 (a) and (b) show the sampled distance (with the polygon of the respective numbers of maximum) of the exemplary minimum that is used to draw body;
Figure 13 (c) and (d) be respectively Figure 13 (a) and grayscale version (b);
Figure 14 has described the divergent-ray with random offset according to exemplary embodiment of the present invention; And
Figure 14 (a) is the grayscale version of Figure 14.
Notice that this patent or application documents comprise at least one width of cloth color accompanying drawing.When request and payment necessary fee, United States Patent Office (USPO) will provide this patent with color drawings or the copy of public announcement of a patent application.The gray scale accompanying drawing of every width of cloth color drawings is provided for illustrative purposes, equally.In the following description, the colour of given accompanying drawing and grayscale version will together be called this accompanying drawing (for example " Fig. 4 " comprises cromogram " Fig. 4 " and corresponding gray-scale map " Fig. 4 (a) " thereof), and promptly accompanying drawing is understood to include all versions of accompanying drawing.
Embodiment
Exemplary embodiment of the present invention is at using the ray emission technology to improve the final render speed of the viewable portion of body.
When drawing body, final render speed and following factor are inversely proportional to: (a) data length---data length is big more in input, and storer that is consumed when drawing it and CPU time are just many more; (b) the texture storage device of video card (texture memory) if physical size and the needed texture storage device of program size ratio---needed texture storage device has surpassed the physical texture memory-size, then can relate to the exchange of texture storage device, this is operation consuming time.In fact, this exchange may take place continually when handling mass data, thereby causes performance sharply to reduce; (c) current time body to be drawn (crop box) size---crop box is more little, need sampling and the polygonal quantity of drawing few more; (d) sharpness of Hui Zhiing (that is the polygonal quantity of using)---sharpness is high more, needs many more polygons; And (e) shade (shading) if use---can use shade, then need 4 times texture storage device.
So, if can optimize one or more in the above-mentioned factor, will improve final render speed.In exemplary embodiment of the present invention, this can realize by the size of optimizing crop box.
In exemplary embodiment of the present invention, can use the ray emission algorithm to calculate the size of crop box.In order to use such exemplary algorithm effectively, need to handle following problem:
A. each display frame amount of radiation of launching.In theory, The more the better, but the ray of emission is many more, and processing speed is slow more;
B.3D the distribution of ray in the space.Ray should cover all surfaces of being paid close attention to.In order to realize this point, in exemplary embodiment of the present invention, for example, can make the layout randomization of ray, so can obtain bigger coverage rate for the ray of equal number.For the zone that needs are noted more, for example, can launch more ray towards them; For the zone that not too should be noted that, can use the ray of smaller amounts; And
C. ray emission result's use (single frames is to multiframe).In exemplary embodiment of the present invention, in each frame, can both collect point of impact result.In an exemplary enforcement, can partly that is in current display frame, use this result, and after the calculating of crop box, abandon; Selectively, for example, can preserve information and be used for subsequent frame, therefore can obtain better result and needn't carry out other calculating to determined number.
For illustrative purposes, use illustrates the present invention as for example exemplary tubulose structure of colon and so on.Expansion for seeing its a fraction of 3D data set at any time to those users also fully is included within the scope of the present invention.
In exemplary embodiment according to the present invention, the 3D display system can be determined the visibility region of the given tubular anatomical structure around user's observation point as be concerned about zone, and does not need to draw the remainder of this tubular structure.For example, the user of virtual observation colon does not check the whole inwall of colonic lumen usually simultaneously in the virtual coloscope inspection.On the contrary, the user only observes the sub-fraction or the segment of inner colon in the time of one.Fig. 1 shows the exemplary endoscope figure of the small pieces of such inside colon.For example, as shown in Figure 2,, can select such segment to show by forming the frame that centers on interest region within the total.Selected segment is full of main viewfinder usually, as shown in Figure 1, so can fully at length watch.So, along with user's observation point moves in whole colonic lumen, unnecessary drafting comprises the complete volumetric data set of whole colon, but only draws the part that the user will watch at any given time point.Needn't draw the user from his current observation point his sightless voxel, thereby greatly optimize system performance and reduced the load on the computational resource.In exemplary embodiment of the present invention, negative load can be reduced to only 3% to 10% of whole scanning, and this is significant optimization.
Like this, in exemplary embodiment according to the present invention, just can use " divergent-ray " method.For example, can construct ray, any position in the 3D model space begins, and finish any other position in the 3D model space.Shown such " ray emission " in Fig. 3 and 4, wherein, Fig. 3 shows the ray among the current endoscope figure that is transmitted into colon, and Fig. 4 then shows the divergent-ray when observing from the side.By check the value of each voxel that ray passes with respect to defined threshold, such example system can obtain the information about any " visibility " at 2.The voxel of the air between the expression internal chamber wall is " sightless ", and ray can pass them.In case arrive first " visible " voxel, just obtained the position of the voxel on the internal chamber wall.Such position is known as " point of impact " sometimes.
In exemplary embodiment of the present invention, can realize the algorithm of such ray emission according to following exemplary false code.
A. the false code that is used for distribute_rays:
Draw in the circulation at each:
Determine projection width and height; // (if change---otherwise see below)
Projection plane is divided into m * n grid, and each grid has (width/m) * (size highly/n); And
Launch a ray from current observation point towards the center of each grid.
Integer m and n for example can equal 5, perhaps for example can adopt other values that are fit to given enforcement.In many exemplary embodiments, projection width and highly be known facts, as for example in any OpenGL program (when the user specifies), thereby its variation always; So, just need in each circulation, all not determine these values under these circumstances.
B. the false code that is used for ray_shooting:
For every ray,
1. from the starting point of ray direction, pick up first voxel along this path towards this root ray;
2. for described voxel, check whether the intensity level of this voxel has surpassed certain threshold level;
If,
3. it is " solid " voxel so, and gets its conduct " point of impact " position, position, returns;
If not,
4. forward to along this path and pick up next voxel, forward 2 to;
5. the voxel that if there is no will pick up (for example ray leaves from this body) then returns;
In exemplary embodiment of the present invention, the direction of ray is exactly the center from current observation point to each grid, and can for example press following setting:
ray.SetStartingPoint(currentViewpoint.GetPosition())；
ray.SetDirection(centerOfGrid-currentViewpoint.GetPosition())；
C. the false code that is used for calculating_bounding_box:
(x, y z), find minimum and maximal value to be respectively Xmin, Xmax, Ymin, Ymax, Zmin and Zmax for all " point of impact " coordinates.So that (Xmin, Ymin is Zmin) as an angle and so that (Zmax) square frame as relative angle is exactly required bounding box for Xmax, Ymax.
So, by using such " divergent-ray " method, in exemplary embodiment of the present invention, system just can for example construct the ray of any amount and send them along any direction from user's current observation point.In these rays some (if not whole words) will finally be hit voxel on the internal chamber wall along their given directions; This has produced one group " point of impact ".The group of such point of impact thereby described from the scope in this visible zone of special observation point.For example in Fig. 3 and 4, the point of impact that is produced is shown as the colored point of yellow or cyan respectively in color drawings, perhaps is shown as white cross and black cross in the gray scale accompanying drawing.The cyan point that shows among Fig. 3 (black cross) has shown the point of impact that is for example generated by the one group of ray that is evenly distributed in the visibility region.Yellow dots will (white cross) has been indicated the point of impact that is used for another group divergent-ray, and described another group divergent-ray only aims at the part of this body, is centered at an end of the center line of exemplary colonic lumen.Owing to can one by one calculate each the distance from the point of impact to user's observation point, so this technology can be used in the visibility frame of dynamically describing from any given observation point.Voxel in such visibility frame is exactly the only voxel that need draw when the user is in this given observation point.The visibility frame for example can have irregularly shaped.In order to be easy to calculate, example system can for example be surrounded the visibility frame by simple " crop box " that is shaped, and " crop box " of described simple shaping for example is cylinder, spheroid, cube, rectangular prism or other simple 3D shapes.
In Fig. 5, further show said method.With reference to described accompanying drawing, in Fig. 5, represent user's observation point by eye icon.From this observation point, exemplary ray can for example be launched on a plurality of directions, and it hits the surface of structure at shown some place.Can be spliced into the rectangular area then, so that within certain user-defined safety allowance, comprise all points of impact.
In exemplary embodiment of the present invention, for example can generate bounding box by the mode of the safety allowance of following regulation:
D. the false code that is used for calculate_bounding_box_safety_margin:
For have the angle (Xmin, Ymin, Zmin) and (Xmax, Ymax, Zmax) bounding box is padded side-play amount to it, makes frame become (Xmin-offset, Ymin-offset, Zmin-offset) and (Xmax-offset, Ymax-offset, Zmax-offset);
Wherein side-play amount can be identical, perhaps can be arranged for each branch in X, Y and the Z direction.
In exemplary embodiment of the present invention, such rectangular area for example can surround the visibility zone about the right wall of tubular structure, as describing among Fig. 5.Similar techniques can for example be applied to left wall, and for this observation point total crop box that produces like this.
In exemplary embodiment according to the present invention, usually, for example, 40 to 50 such rays that spread all over the current visual field of user just can be collected the enough information about tubular structure surface placement, make formation visibility zone.In exemplary embodiment of the present invention, the quantity of the ray of emission is adjustable.So, the ray of emission is many more, and the result is just good more, but calculates slow more.So, in exemplary embodiment of the present invention, the quantity of the ray of emission just can be got suitable value in given range, thus these two factors of balance, that is computing velocity and the needed degree of accuracy of optimization crop box.
Be used for the above-mentioned false code of calculate_bounding_box, wherein stated with minor function { for all " point of impact " coordinate (x, y, z) }, if " point of impact " is not only from present frame, but also from former several frames, then in exemplary embodiment of the present invention, accurate computation bound frame still.In fact, if preserved enough information from former frame, result even can be better in the exemplary embodiment then.
In exemplary embodiment of the present invention, available by the point of impact that used as getting off from former frame:
E. the false code that is used for the point of impact before follow-up frame uses:
Show circulation for each,
Hit_points=ShootRays; // as above
Hit_points_pool.add (hit_points); // the new point of impact is added in " pond ",
// Storage
All points of impact are determined crop box among the use hit_points_pool;
// only used current hit_points in the past,
Hit_points in the // previous cycles is deleted,
// and never reuse
In exemplary embodiment of the present invention, hit_points_pool for example can store not only from current but also from before (one or several) round-robin hit_points.Like this, in each circulation, be used for determining that the quantity of the hit_points of crop box just can be greater than the quantity of the ray of reality ejaculation; Thereby all hit_points can both for example store among the hit_points_pool, and reuse in the circulation afterwards.
As mentioned above, by collecting information, in exemplary embodiment of the present invention, can utilize the coordinate of such point of impact to surround their whole (axle is aimed at) crop boxs to produce about the point of impact.This can be in the definition of given given viewpoint for visible zone of user or the zone paid close attention to.Such crop box for example can be used for reducing the actual amount of the whole body that need draw at any given time, as mentioned above.Notice that for many 3D data sets, desirable crop box may not be (that is aim at x, y and the z axle of this body) that axle is aimed at, but can be for example in given given viewpoint with the cone (viewing frustrum) aligning.Such aligning has further reduced the size of crop box, but calculates complexity more for being plotted in.Fig. 8 (a)-(d) has described the difference between the crop box that crop box that axle aims at and the cone aim at.So, feasible and be desirable when the crop box of freely aiming in exemplary embodiment of the present invention, crop box can for example be that the cone is aimed at, or any other mode that given data set and available computational resources are fit to is aimed at.
With reference to figure 8, shown so exemplary crop box of freely aiming at.Fig. 8 (a) has described the exemplary cone about the given given viewpoint of whole exemplary colon body.As can seeing, such awl does not have specific aiming at naturally with the axle of this body.Fig. 8 (b) has described the exemplary hit point that obtains as mentioned above.Fig. 8 (c) has described to comprise the crop box that the exemplary axon of these points of impact is aimed at.As can seeing, crop box has the exceptional space that gibberish occurred.But these voxels will be drawn in showing circulation.Fig. 8 (d) has described the crop box that the exemplary cone is aimed at, wherein the crop box direction of aiming at viewpoint direction and being orthogonal to this direction vector in 3d space.As can seeing, such crop box " naturally " matches with the shape of these data, thereby and can diminish significantly, yet, in order to indicate the voxel that comprises within it, in exemplary embodiment of the present invention, example system may need to carry out the coordinate transform of computation-intensive.
In the exemplary embodiment, the size of crop box can be significantly less than the body of whole structure to be analyzed.For example, in exemplary embodiment of the present invention, it can be to be used for below 5% of initial body that colonoscopy is used.Therefore, can greatly improve render speed.
As mentioned above, render speed depends on many factors.Fig. 9-13 show sampled distance (that is for draw this said three-dimensional body that is used for sampling again perpendicular to the distance between the polygon of direction of observation), relation between needs polygonal quantity, rendering quality and the crop box drawn.
The left part of each among Fig. 9-13 (that is part of mark (a) and accompanying drawing (c)) exhibit textural polygon, right side part (that is those parts of mark (b) and accompanying drawing (d)) then only shows polygonal limit.In any given moment, in fact shown all polygonal scopes constitute cube shaped, this has reflected the following fact: polygonal size is determined by crop box, described crop box calculated before this stage, that is, show in the circulation at each, calculate crop box before and then showing.So in fact, polygon has been indicated the shape of crop box.
Fig. 9 produces by intentionally specifying very large sampled distance, and it causes using between sampling period more considerably less polygon.This has provided low-down sharpness.The polygonal quantity that shows among Fig. 9 only is about 4 or 5.
Sampled distance reduces in Figure 10, and therefore polygonal quantity increases.Yet remain insignificant according to this value image.Figure 11 and 12 described sampled distance further reduce (and corresponding sampled distance increase) thus effect and provided higher sharpness, and as a result of, the shape of inner chamber appears can discern more.Yet polygonal quantity sharply increases.
At last, in Figure 13, seen best picture quality, and these accompanying drawings are to use thousands of polygons to generate.Polygonal limit is so closely each other so that among the figure on the right (that is 13 (b) and (d)) they appear to have connected into face.
A kind of clumsy method that acquisition can be surrounded the crop box of all visible voxels is, launches and some rays that the quantity of the pixel that is used to show equates, so just covered whole screen area.Yet if screen area is for example 512 * 512 pixels, this need launch approximate 512 * 512=262,144 rays.Because must handle the pixel and the ray of related quantity, such method usually is unpractical.
So, in exemplary embodiment of the present invention, just can launch one group of ray, its resolution for example is enough to catch the shape of visible borders.Such ray groups shows with cyan (black cross) in Fig. 3.
As from can seeing the Fig. 3 and 6 that has described exemplary colon, the depth capacity great majority at certain observation point place manifest at the rear portion of center line usually.This is because the user generally sees in colon in endoscope figure, perhaps points to caecum, perhaps points to rectum.So, the equally distributed ray (being shown as cyan ray or black cross in Fig. 3 and 6) that spreads all over the emission of colon body will can not hit the border farthest of visible voxel.If the distance between the ray (can be described as their " resolution ") is for example greater than the diameter (in this case) in the colonic lumen at image rear portion, Fa She ray may all return the point of impact of too close observation point so, to such an extent as to do not comprise the aft section of colonic lumen in crop box.So, in Fig. 6, the aft section of tubular structure just is not shown, and black picture element filled should blank.In order to remedy this, in exemplary embodiment of the present invention, can check center line (or other zones of the part correlation connection of the visibility frame omitted of first group of low resolution ray of known and emission), make can with respect to screen area determine " pipe " visible part other end where.
In exemplary embodiment of the present invention, for example, this can be by following realization:
The false code that is used for distribute_rays in front:
After step 3,
4. lead to where come to determine the zone paid close attention to by finding out center line;
5. the part that will comprise the projection plane of this interest region further is divided into littler grid; And
6. launch a ray towards the center of each grid.
For example step (4) can be implemented as follows.Because in exemplary embodiment of the present invention, exemplary process can have position and its position on center line and the shape of center line of current observation point, so this program for example can be checked through simply along current direction increment and leave on the center line N centimetre point, till such point is no longer visible; On projection plane, for example can determine the relevant position of last visible point then:
Be used to determine the zone paid close attention to exemplary pseudo-code (top step 4):
1. obtain current observation point position P
0
2. obtain current observation point position P
0Relative position on center line (according to initial what centimetres that leave) from center line
3. obtain centerline points P
i, it leaves (n * i) centimetre (supposition n=5cm) from current observation point;
4. pass through from P
0To P
iDivergent-ray is checked P
0And P
iWhether mutually as seen:
If at P
0And P
iBetween exist the point of impact (to this means that ray arrives P
iHit before), P so
0And P
iBe sightless; Return P
(i-1)
Otherwise: i=i+1; Forward 3 to;
Step (5) can for example be implemented as follows:
Be used for the exemplary pseudo-code that grid divides again (top step 5):
For the last visible point that calculated in the former step,
1. obtain this projection on projection plane;
2. obtain the rectangular area of the size 1/m of whole projection plane, it is centered at this point (m=5 in fact for example is set) on projection plane; And
3. this rectangular area is divided into m * m grid (for m=5,25 grids).
Like this, in exemplary embodiment of the present invention, for example, system can launch the other ray that is centered at the center line end, so that use above-described ray emission method to fill the part of omitting, but has interval between much bigger resolution or the much smaller ray.Figure 7 illustrates the result of this method, wherein, tubular structure no longer has the part of omission because second group of ray (in Fig. 7, illustrating) with yellow or white cross thus obtained enough points of impact to catch its shape and fully it to be enclosed in the crop box along actual boundary.
The situation of describing among consideration Fig. 6, in replacement exemplary embodiment of the present invention, when attempting to catch the scope of desired crop box better, divergent-ray may be otiose unchangeably on fixing direction.On the contrary, in such embodiments, for example can use random offset to carry out ray emission so that between the point of impact apart from disunity.This can eliminate above-described divergent-ray " low resolution " problem.Figure 14 illustrates such technology, wherein, in each circulation, numeral 1,2 ..., 6 expression circulations 1,2 respectively ..., the ray of being launched in each of 6 all has different randomized side-play amounts at every turn.
So, with reference to Figure 14, use as above the exemplary pseudo-code that is used for distribute_rays providing, exemplary embodiment as can be not only towards each grid ray of heart emission that hits really, but also direction that for example can every ray of randomization, so that the direction of ray (dx, dy) become (dx+random_offset, dy+random_offset).
Use such example technique, it is identical that the total quantity of the ray of emission keeps, but the ray in the consecutive frame does not send along same path.Compare with the ray method that uses fixed-direction, this method thereby can for example cover the viewing area more all sidedly, and can eliminate needs in the exemplary embodiment to second group of (" more high resolving power ") ray of concentrating more, such as shown in Figure 7, it is transmitted in the part of this body, wherein, known boundaries has little slit (with respect to distance between the ray of first group of ray), but has big+Z coordinate (that is from observation point, it extends to far place in the screen).
Example system
With the form of the software that moves on the data processor, with the form of the hardware of one or more special chip forms or with the form of above-mentioned any combination, can implement the present invention.Example system can comprise that for example three-dimensional display, data processor, mapping show one or more interfaces of control command and function, one or more storer or memory storage and graphic process unit and related system alternately.For example, the system that Dextroscope that the Volume Interactions Pte Ltd by Singapore of operation RadioDexter software makes and Dextrobeam system come to this can easily implement method of the present invention in described system.
Exemplary embodiment of the present invention can be implemented as the modular software program of the instruction that can be carried out by suitable data processor, as known in the art or may be known, so realizes the preferred embodiments of the present invention.Exemplary software programs for example can be stored on hard disk drive, flash memories, memory stick, optical storage medium or other data storage devices, as known in the art or may be known.When such program during by the CPU of suitable data processor visit and operation, in exemplary embodiment of the present invention, it can carry out above-mentioned method, shows 3D computer model or a plurality of model of tubular structure in the 3D data presentation system.
Although with reference to its one or more exemplary embodiments the present invention has been described, but the present invention is not limited to this, but subsidiary claim should be interpreted as not only comprising shown particular form of the present invention and remodeling, is not deviated from those of true scope of the present invention by those skilled in the art's design but also further comprise picture.
Claims (17)
1. method that is used for the dynamic demonstration of optimization 3D data set, this method comprises:
Determine the border of the relevant portion of 3D data set from current observation point;
The described relevant portion that shows described 3D data set; And
When the changes in coordinates of described current observation point, repeat describedly to determine and described procedure for displaying.
2. the relevant portion of the method for claim 1, wherein described 3D data set is the endoscope figure of tubular structure.
3. method as claimed in claim 2, wherein, by realizing described definite border to described tubular structure around the inwall divergent-ray from current observation point.
4. the relevant portion of the method for claim 1, wherein described 3D data set is the endoscope figure of colon.
5. method as claimed in claim 4 wherein, realizes described definite border to described tubular structure around the inwall divergent-ray by the current observation point from center line.
6. method as claimed in claim 3, wherein, described ray is the observation point emission from the center line of described tubular structure, and the described ray that distributes is so that cover visibility region.
7. method as claimed in claim 6, wherein, described ray evenly distributes in the scope of described visibility region.
8. method as claimed in claim 6, wherein, the direction of launching described ray comprises random component.
9. method as claimed in claim 3, wherein:
To first area in launching first group ray from current observation point with first resolution within the tubular structure; And
Launch second group ray from described current observation point towards second area with second resolution,
Wherein said second area is the subclass of described first area.
10. method as claimed in claim 9, wherein, described second area is confirmed as described first group of ray sampling may not enough zone.
11. method as claimed in claim 9 wherein, is checked around the zone of the described tubular structure of following direction, has found to have the visible voxel apart from observation point ultimate range on described direction, comes to determine defined zone thus.
12. method as claimed in claim 9 wherein, becomes invisible part by checking described center line in current scene, determine defined zone.
13. as claim 3 or 9 described methods, wherein, in tubular structure along each some place of the divergent-ray of center line, described ray of emission from each of two observation point of expression position of human eye.
14. a computer program comprises:
But the computing machine working medium, it has the computer-readable program code means that comprises within it, and the described computer-readable program code means in the described computer program comprises such device so that computing machine:
Determine the border of the relevant portion of 3D data set from current observation point;
The described relevant portion that shows described 3D data set; And
When the changes in coordinates of described current observation point, repeat describedly to determine and described procedure for displaying.
15. a machine-readable program storage device, the actual program that comprises the executable instruction of described machine is used for the method for the dynamic demonstration of optimization 3D data set with execution, and described method comprises:
Determine the border of the relevant portion of 3D data set from current observation point;
The described relevant portion that shows described 3D data set; And
When the changes in coordinates of described current observation point, repeat describedly to determine and described procedure for displaying.
16. computer program as claimed in claim 14, wherein, described device further makes computing machine:
To first area launching first group ray from current observation point with first resolution within the tubular structure; And
Launch second group ray from described current observation point towards second area with second resolution, wherein said second area is the subclass of described first area.
17. program storage device as claimed in claim 15, wherein, described method further comprises:
To first area launching first group ray from current observation point with first resolution within the tubular structure; And
Launch second group ray from described current observation point towards second area with second resolution,
Wherein said second area is the subclass of described first area.
Applications Claiming Priority (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US51704303P | 2003-11-03 | 2003-11-03 | |
| US60/517,043 | 2003-11-03 | ||
| US60/516,998 | 2003-11-03 | ||
| US60/562,100 | 2004-04-14 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN1879128A true CN1879128A (en) | 2006-12-13 |
Family
ID=37484943
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2004800328701A Pending CN1879128A (en) | 2003-11-03 | 2004-11-03 | Dynamic crop box determination for optimized display of a tube-like structure in endoscopic view |
| CNA2004800327041A Pending CN1875381A (en) | 2003-11-03 | 2004-11-03 | System and methods for screening a luminal organ-lumen viewer |
Family Applications After (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2004800327041A Pending CN1875381A (en) | 2003-11-03 | 2004-11-03 | System and methods for screening a luminal organ-lumen viewer |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN1879128A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108335284A (en) * | 2018-01-09 | 2018-07-27 | 北京理工大学 | A kind of coronary artery vessel centerline matching process and system |
| CN109887064A (en) * | 2013-07-23 | 2019-06-14 | 玛口外科股份有限公司 | The method and system generated for radioscopic image |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5421828B2 (en) * | 2010-03-17 | 2014-02-19 | 富士フイルム株式会社 | Endoscope observation support system, endoscope observation support device, operation method thereof, and program |
| JP5551955B2 (en) * | 2010-03-31 | 2014-07-16 | 富士フイルム株式会社 | Projection image generation apparatus, method, and program |
| JP6391922B2 (en) * | 2012-08-08 | 2018-09-19 | キヤノンメディカルシステムズ株式会社 | Medical image diagnostic apparatus, image processing apparatus, and image processing method |
| US9510757B2 (en) | 2014-05-07 | 2016-12-06 | Align Technology, Inc. | Identification of areas of interest during intraoral scans |
| CN106413516B (en) * | 2014-07-30 | 2018-01-23 | 奥林巴斯株式会社 | Image processing apparatus |
-
2004
- 2004-11-03 CN CNA2004800328701A patent/CN1879128A/en active Pending
- 2004-11-03 CN CNA2004800327041A patent/CN1875381A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109887064A (en) * | 2013-07-23 | 2019-06-14 | 玛口外科股份有限公司 | The method and system generated for radioscopic image |
| CN109887064B (en) * | 2013-07-23 | 2023-10-27 | 玛口外科股份有限公司 | Method and system for X-ray image generation |
| CN108335284A (en) * | 2018-01-09 | 2018-07-27 | 北京理工大学 | A kind of coronary artery vessel centerline matching process and system |
| CN108335284B (en) * | 2018-01-09 | 2022-06-28 | 北京理工大学 | Coronary vessel center line matching method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1875381A (en) | 2006-12-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20050116957A1 (en) | Dynamic crop box determination for optimized display of a tube-like structure in endoscopic view ("crop box") | |
| CN1054921C (en) | Virtual internal cavity inspection system | |
| JP4518470B2 (en) | Automatic navigation for virtual endoscopy | |
| US7304644B2 (en) | System and method for performing a virtual endoscopy | |
| EP3613019B1 (en) | System and method for combining 3d images in color | |
| JP5976003B2 (en) | Method and system for visualizing volume datasets | |
| CN1759812A (en) | Ultrasonic diagnostic equipment and image processing method | |
| CN1860991A (en) | Real-time virtual endoscopy | |
| US20090003668A1 (en) | Image processing method, image processing program, and image processing device | |
| JP2006055213A (en) | Image processor and program | |
| CN1856805A (en) | Generation of motion blur | |
| US7576741B2 (en) | Method, computer program product, and device for processing projection images | |
| CN1879128A (en) | Dynamic crop box determination for optimized display of a tube-like structure in endoscopic view | |
| CN1827046A (en) | Image processing apparatus | |
| CN100583161C (en) | Method for depicting an object displayed in a volume data set | |
| JP7180123B2 (en) | Medical image processing apparatus, medical image processing method, program, and data creation method | |
| US10891776B2 (en) | Imaging system and method | |
| CN108876783A (en) | Image interfusion method and system, Medical Devices and image co-registration terminal | |
| JP7114362B2 (en) | MEDICAL IMAGE PROCESSING APPARATUS AND MEDICAL IMAGE PROCESSING PROGRAM | |
| CN101067870A (en) | High light hot spot eliminating method using for visual convex shell drawing and device thereof | |
| Zhang et al. | Real-time visualization of 4D cardiac MR images using graphics processing units | |
| KR101977650B1 (en) | Medical Image Processing Apparatus Using Augmented Reality and Medical Image Processing Method Using The Same | |
| JP2019205796A (en) | Medical image processing apparatus, medical image processing method, program, and MPR image generation method | |
| Harris | Display of multidimensional biomedical image information | |
| Turlington et al. | Improved techniques for fast sliding thin-slab volume visualization |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| 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 |
Open date: 20061213 |