CN106530382A - Data processing method and system for medical three-dimensional image - Google Patents
Data processing method and system for medical three-dimensional image Download PDFInfo
- Publication number
- CN106530382A CN106530382A CN201611126881.7A CN201611126881A CN106530382A CN 106530382 A CN106530382 A CN 106530382A CN 201611126881 A CN201611126881 A CN 201611126881A CN 106530382 A CN106530382 A CN 106530382A
- Authority
- CN
- China
- Prior art keywords
- data
- cgpuspeedui
- class
- medical
- video card
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T15/00—3D [Three Dimensional] image rendering
- G06T15/005—General purpose rendering architectures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F8/00—Arrangements for software engineering
- G06F8/30—Creation or generation of source code
- G06F8/34—Graphical or visual programming
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T1/00—General purpose image data processing
- G06T1/20—Processor architectures; Processor configuration, e.g. pipelining
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2215/00—Indexing scheme for image rendering
- G06T2215/06—Curved planar reformation of 3D line structures
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Computer Graphics (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
The invention is applicable to the technical field of medical image data processing, and provides a data processing method for a medical three-dimensional image. The method comprises the following steps: A, inputted data are received; B, after a graphics processing unit is initialized, data traced by screen pixel light are copied to a Device end; C, a function is called to carry out light tracing sampling synthesizing calculation at the Device end; and D, screen pixel result data are coped to a Host end. The invention correspondingly provides a data processing system for the medical three-dimensional image for realizing the above method. Thus, the three-dimensional visual processing speed can be quickened, and transplantation of a large number of codes can be avoided.
Description
Technical field
The present invention relates to medical image processing technology field, more particularly to a kind of data processing of medical threedimensional images
Method and system.
Background technology
In clinical practice in addition to proposing higher requirement to three-dimensional visualization reconstructed image quality, to Interactive Visualization speed
Degree it is also proposed higher requirement.High-quality 3-D view can be generated based on light projection method, disclosure satisfy that clinic is examined
Disconnected needs.But the prominent shortcoming of its one is exactly to draw speed slowly, if not having a kind of effective accelerated method to draw which
If speed is improved, it is difficult to adapt to the requirement of practical application and the development of medical image technology.Light projection method propose with
Afterwards, the research of corresponding speed technology just never stopped.At present, the Software Acceleration Techniques for being directed to light projection are main
Have following several：Box technology (bounding box), uniformly subdivision (space subdivision) technology are related to light
Property (ray coherence) technology.
Above-mentioned some accelerated methods, are mainly all based on ignoring the thought of invalid computation to reduce time loss,
Should say and all achieve good acceleration effect, but the medical image three-dimensional visualization increasing for data volume, still very
Difficulty reaches real-time display effect.
In summary, prior art there will naturally be inconvenience and defect in actual use, it is therefore necessary to being improved.
The content of the invention
For above-mentioned defect, it is an object of the invention to provide a kind of data processing method of medical threedimensional images and being
System, which can accelerate the processing speed of three-dimensional visualization, and avoid substantial amounts of code migrating.
To achieve these goals, the present invention provides a kind of data processing method of medical threedimensional images, including following step
Suddenly：
A, receiving data input；
After B, graphic process unit initialization, described in copy screen pixels ray tracing, data are to Device ends；
C, call function carry out ray tracing sampling composite calulation at Device ends；
D, copy screen pixels result data are to Host ends.
The data processing method of medical threedimensional images of the invention, step B are further included：
B1, establishment interface derive class CGPUSpeedUI；
B2, under VS2008 environment by the interface derivation class be compiled into corresponding lib files；
B3, the data that priming dose calculating needs in the CPU programs of host ends；
B4, the data needed the Rapid Dose Calculation by the member function of calling interface class CGPUSpeedUI are answered
Make in video card internal memory；
B5, device end core program performing terminate after again by member's letter of calling interface class CGPUSpeedUI
Number is by the data duplication for calculating in same main memory.
The data processing method of medical threedimensional images of the invention, in step B4, by video card internal memory only
Reading it is stated that be data texturing, by the read-write data assertion in video card internal memory be global variable.
The present invention also correspondingly provides a kind of data handling system of medical threedimensional images, including：
Data reception module, is input into for receiving data；
Graphic process unit, for data described in screen pixels ray tracing are copied after initialization to Device ends；
Processing module is calculated, call function carries out ray tracing sampling composite calulation at Device ends；
Data transmission module, copy screen pixels result data is to Host ends.
The data handling system of medical threedimensional images of the invention, the graphic process unit are further used for：
Creating interface derives class CGPUSpeedUI；
Interface derivation class CGPUSpeedUI is compiled into into corresponding lib files under VS2008 environment；
In the CPU programs of host ends, priming dose calculates the data for needing；
The data duplication that the Rapid Dose Calculation needs is arrived by the member function of calling interface class CGPUSpeedUI
In video card internal memory；
Will by the member function of calling interface class CGPUSpeedUI after device end core program performings terminate
The data duplication for calculating is in same main memory.
The data handling system of medical threedimensional images of the invention, the graphic process unit are further used for video card
Read-only data in internal memory is claimed as data texturing, is global variable by the read-write data assertion in video card internal memory.
The present invention is based on NVIDIA video card general-purpose computations framework CUDA technologies, and GPU speed technologies are applied to medical 3 D
During visualization body shows, three-dimensional visualization is rebuild speed and is greatly improved.Meanwhile, using MFC extended dynamics storehouse and
Class technology is derived, a large amount of code migrating work are avoided in architecture design.
Description of the drawings
Fig. 1 is the data handling system structural representation of the medical threedimensional images of the present invention；
Fig. 2 is the architecture diagram of one embodiment of the invention；
Fig. 3 is the data processing method flow chart of the medical threedimensional images of one embodiment of the invention.
Specific embodiment
In order that the objects, technical solutions and advantages of the present invention become more apparent, it is below in conjunction with drawings and Examples, right
The present invention is further elaborated.It should be appreciated that specific embodiment described herein is only to explain the present invention, and
It is not used in the restriction present invention.
The present invention be based on NVIDIA video card general-purpose computations framework CUDA (parallel computation architecture platform) technologies, by GPU plus
Fast technology is applied to medical 3 D visualization body and shows, makes three-dimensional visualization reconstruction speed be greatly improved.Meanwhile, profit
With MFC extended dynamics storehouse and derivation class technology, a large amount of code migratings work are avoided in architecture design.And for specific
GeForce 9600GT. video cards, in process of the test, are counted with the calculating time to the Thread Count that GPU is performed simultaneously, are found
The time of calculating minimum optimum Thread Count.
In the present invention, if no special instructions, involved VS6 and VS2008 are the original of computer disposal field and contain
Justice.
Referring to Fig. 1, the invention provides a kind of data handling system of medical threedimensional images, the system 100 includes：
Data reception module 10, is input into for receiving data.
Graphic process unit 20, for screen pixels ray tracing desired data is copied after initialization to Device ends.
Processing module 30 is calculated, call function carries out ray tracing sampling composite calulation at Device ends.
Data transmission module 40, copy screen pixels ray tracing result data is to Host ends.
In concrete application, by the ray tracing process of light quantum mechanics Composite Display pixel for each pixel is only
Vertical, the ray tracing process of each pixel can be completed by a thread in GPU (graphic process unit), with very
Good parallel computation characteristic, based on this, the invention provides acceleration is processed.
Screen pixels after discrete processes, for any discrete point of screen correspondence pixel color value cλ(um, vn) calculating
Can independently carry out in display chip performs least unit thread.Accelerator is the meter based on the process of graphic process unit 20
Calculate and improve.Specifically, the present invention is pre-created interface and derives class CGPUSpeedUI, and by the interface under VS2008 environment
Derive class CGPUSpeedUI and be compiled into corresponding lib files.When system works, preferentially can prepare in the CPU programs of host ends
The data that Rapid Dose Calculation needs, and the Rapid Dose Calculation is needed by the member function of calling interface class CGPUSpeedUI
The data duplication wanted in video card internal memory, by the calling interface class after device end core program performings terminate
The member function of CGPUSpeedUI is by the data duplication for calculating in same main memory.
In above process, these data being copied in video card internal memory include needs in device ends kernel programs
Need some arrays and variable of definition.In CUDA memory managements, texture memory has read only attribute, and global memory can enter
Row read-write operation.But texture memory space has cache, texture is picked up only in cache-miss, just can be from equipment
Data are read in internal memory, so texture memory is compared with global memory, with higher reading efficiency.For above characteristic, I
Read-only array is managed using texture memory mode, for read-write array is claimed as global variable, can carry significantly whereby
The processing speed of high data.
Preferably, with reference to Fig. 2, class CGPUSpeedUI is derived due to invention creates an interface, will under VS2008
Which is compiled into corresponding lib files.VS6 engineerings are compiled by the lib files.The kernel programs at so device ends can
Performed with the dynamic base under VS2008, it is not necessary to substantial amounts of code migrating work is done to original engineering.
Referring to Fig. 3, the invention provides a kind of data processing method of medical threedimensional images, which can pass through such as Fig. 1 institutes
The system 100 shown realizes that the method includes：
Step S301, receiving data input.
Step S302, after graphic process unit initialization, copy screen pixels ray tracing desired data is to Device ends.
Step S303, call function carry out ray tracing sampling composite calulation at Device ends.
Step S304, copy screen pixels ray tracing result data is to Host ends.
Specifically, in step s 302, the present invention is pre-created interface and derives class CGPUSpeedUI, and in VS2008 rings
Interface derivation class CGPUSpeedUI is compiled into into corresponding lib files under border.When system works, can be preferentially in host
In the CPU programs of end, priming dose calculates the data for needing, and will by the member function of calling interface class CGPUSpeedUI
The data duplication that the Rapid Dose Calculation needs passes through the tune in video card internal memory, after device end core program performings terminate
With the member function of interface class CGPUSpeedUI by the data duplication for calculating in same main memory.
In above process, these data being copied in video card internal memory include needs in device ends kernel programs
Need some arrays and variable of definition.In CUDA memory managements, texture memory has read only attribute, and global memory can enter
Row read-write operation.But texture memory space has cache, texture is picked up only in cache-miss, just can be from equipment
Data are read in internal memory, so texture memory is compared with global memory, with higher reading efficiency.For above characteristic, I
Read-only array is managed using texture memory mode, for read-write array is claimed as global variable, can carry significantly whereby
The processing speed of high data.
In sum, the present invention is based on NVIDIA video card general-purpose computations framework CUDA technologies, and GPU speed technologies are applied to
During medical 3 D visualization body shows, three-dimensional visualization is rebuild speed and is greatly improved.Meanwhile, extended using MFC
Dynamic base and derivation class technology, avoid a large amount of code migratings work in architecture design.
Certainly, the present invention can also have other various embodiments, in the case of without departing substantially from spirit of the invention and its essence, ripe
Know those skilled in the art and work as and various corresponding changes and deformation, but these corresponding changes and change can be made according to the present invention
Shape should all belong to the protection domain of appended claims of the invention.
Claims (6)
1. a kind of data processing method of medical threedimensional images, it is characterised in that comprise the following steps：
A, receiving data input；
After B, graphic process unit initialization, copy screen pixels ray tracing desired data is to Device ends；
C, call function carry out ray tracing sampling composite calulation at Device ends；
D, copy screen pixels ray tracing result data are to Host ends.
2. the data processing method of medical threedimensional images according to claim 1, it is characterised in that step B enters
Step includes：
B1, establishment interface derive class CGPUSpeedUI；
B2, under VS2008 environment by the interface derivation class be compiled into corresponding lib files；
B3, the data that priming dose calculating needs in the CPU programs of host ends；
B4, the data duplication that the Rapid Dose Calculation needs is arrived by the member function of calling interface class CGPUSpeedUI
In video card internal memory；
B5, device end core program performing again will by the member function of calling interface class CGPUSpeedUI after terminating
The data duplication for calculating is in same main memory.
3. the data processing method of medical threedimensional images according to claim 2, it is characterised in that in step B4,
Read-only data in video card internal memory is claimed as into data texturing, is global variable by the read-write data assertion in video card internal memory.
4. a kind of data handling system of medical threedimensional images, it is characterised in that include：
Data reception module, is input into for receiving data；
Graphic process unit, for data described in screen pixels ray tracing are copied after initialization to Device ends；
Processing module is calculated, call function carries out ray tracing sampling composite calulation at Device ends；
Data transmission module, copy screen pixels result data is to Host ends.
5. the data handling system of medical threedimensional images according to claim 4, it is characterised in that the graphic process unit
It is further used for：
Creating interface derives class CGPUSpeedUI；
Interface derivation class CGPUSpeedUI is compiled into into corresponding 1ib files under VS2008 environment；
In the CPU programs of host ends, priming dose calculates the data for needing；
The data duplication needed the Rapid Dose Calculation by the member function of calling interface class CGPUSpeedUI is to video card
In internal memory；
To be calculated by the member function of calling interface class CGPUSpeedUI after device end core program performings terminate
Good data duplication is in same main memory.
6. the data handling system of medical threedimensional images according to claim 5, it is characterised in that the graphic process unit
It is further used for for the read-only data in video card internal memory being claimed as data texturing, by the read-write data assertion in video card internal memory is
Global variable.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611126881.7A CN106530382A (en) | 2016-12-09 | 2016-12-09 | Data processing method and system for medical three-dimensional image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611126881.7A CN106530382A (en) | 2016-12-09 | 2016-12-09 | Data processing method and system for medical three-dimensional image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106530382A true CN106530382A (en) | 2017-03-22 |
Family
ID=58342773
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611126881.7A Pending CN106530382A (en) | 2016-12-09 | 2016-12-09 | Data processing method and system for medical three-dimensional image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106530382A (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103106685A (en) * | 2013-01-16 | 2013-05-15 | 东北大学 | Abdominal viscera three dimension visualization method based on graphic processing unit (GPU) |
| US20160005218A1 (en) * | 2014-07-01 | 2016-01-07 | Kabushiki Kaisha Toshiba | Image rendering apparatus and method |
-
2016
- 2016-12-09 CN CN201611126881.7A patent/CN106530382A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103106685A (en) * | 2013-01-16 | 2013-05-15 | 东北大学 | Abdominal viscera three dimension visualization method based on graphic processing unit (GPU) |
| US20160005218A1 (en) * | 2014-07-01 | 2016-01-07 | Kabushiki Kaisha Toshiba | Image rendering apparatus and method |
Non-Patent Citations (2)
| Title |
|---|
| 王玉: "三维放射治疗计划***仿真建模研究与实现", 《万方在线学位论文》 * |
| 王玉等: "GPU加速技术在治疗计划***剂量计算中的应用", 《东北大学学报(自然科学版)》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110084875B (en) | Using a compute shader as a front-end for a vertex shader | |
| EP2697778B1 (en) | Pixel value compaction for graphics processing | |
| CN111815755A (en) | Method and device for determining shielded area of virtual object and terminal equipment | |
| Brabec et al. | Shadow volumes on programmable graphics hardware | |
| US8310484B2 (en) | Efficient processing of operator graphs representing three-dimensional character animation | |
| US7847798B1 (en) | Clipping and culling artwork using a graphics processing unit | |
| CN103310484B (en) | Computed tomography (CT) image rebuilding accelerating method based on compute unified device architecture (CUDA) | |
| US11727632B2 (en) | Shader binding management in ray tracing | |
| CN115509764B (en) | Real-time rendering multi-GPU parallel scheduling method and device and memory | |
| US10580198B2 (en) | System and method for optimized sparse volume rendering | |
| US8264503B1 (en) | Batch early engagement rules graphics processing | |
| CN106204413A (en) | The figure utilizing layering accelerating structure renders | |
| KR101499124B1 (en) | Method and apparratus of image processing using shared memory | |
| CN117237502A (en) | Three-dimensional rendering method, device, equipment and medium | |
| Liu et al. | Parallel Marching Blocks: A Practical Isosurfacing Algorithm for Large Data on Many‐Core Architectures | |
| CN106530382A (en) | Data processing method and system for medical three-dimensional image | |
| Qiu et al. | Parallel fast pencil drawing generation algorithm based on GPU | |
| Lee et al. | A bimodal empty space skipping of ray casting for terrain data | |
| Wang et al. | LVDIF: a framework for real-time interaction with large volume data | |
| CN106384377B (en) | Method and device for volume rendering of medical data | |
| JP6802129B2 (en) | Information processing equipment, methods and programs | |
| Fu et al. | Dynamic shadow rendering with shadow volume optimization | |
| Kumar et al. | Gpu-accelerated interactive visualization of 3D volumetric data using CUDA | |
| Al-Oraiqat et al. | Specialized Computer systems for environment visualization | |
| JP2666353B2 (en) | High-speed video generation by ray tracing |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170322 |