Open Journal of Applied Sciences, 2013, 3, 32-35
Published Online March 2013(http://www.scirp.org/journal/ojapps)
Copyright © 2013 SciRes. OJAppS
Two-Dimensional Medical Image 3D Visualization System’s
Realization
Ai Ting1, Li Zhe1, Miao Yu2
1Engineering and Technology College of Changchun Department of Electronics
2Science and Technology University of Changchun Department of computer
Email: aiting121@sina.com
Received 2012
ABSTRACT
With the development of virtual reality application in the medical field, t wo-dimensional medical image of the
three -dimensional visualization technology made possible. Surgery gets into minimally invasive operation microscopy
Era, and gradually becomes a new research hotspot. This paper studies the realization of two-dimensional medical im-
age 3D reconstruction visualization system method, and the overall process and management module. Using the main
technology of VTK (The Visualization Toolkit) to achieve a two-dimensional medical image three-dimensional visua-
lization system, which can help the physician to obtain help clinical diagnosis Information and play an important role in
treatment, accurate positioning in diseased tissue and tumor early diagnosis.
Keywords: VTK; Visualization in Scientific Computing; 3D Reconstruction; Virtual Endoscopy
1. Introduction
In the medical field, virtual reality technology and mod-
ern surgical operation combining the virtual operation
system for surgical operation, operation guidance and
assessment of teaching and provides a new mode. It
gradually shows its great potential and prospects. With
the development of medical imaging technology, Virtual
operation scene creation is the main organs of the human
body 3D modeling, computer tomography (CT), magnetic
resonance imaging (MRI) and other devices to obtain
patient two-dimensional image information for 3D recon-
struction, direct generation for virtual operation system
geometric model so that the image can intuitively show
the internal body tissues with complex structures, help
doctors diagnosis and guide the operation, the auxiliary
means can make up the deficiency in imaging equipment,
can provide users with a realistic 3D medical image,
which is convenient for doctors from much angle, much
level were observed and analyzed, and could allow doc-
tors to participate effectively in data processing and anal-
ysis.
2. Medical Image Visualization Technology
Medical image visualization is the transformation from
CT, MRI and other digital imaging technology to obtain
the information of human body visually on the computer
performance for the three-dimensional effect, so as to
provide with tradition al method to obtain structural in-
formation. In the medicine fie ld , the information of hu-
man body is obtained by CT, MRI and other digital im-
aging technology which to be visually on the computer
performance for the 3D effect. X ray tomography image
CT and magnetic resonance images MRI in clinical ap-
plication will produces a large amount of data such as
Figure 1, which can effectively display the information
contained in these data, and make full use of these data,
whi c h has a wide range of applications and research val-
ue in the medical field .
Figure 1. med ical image data.
2.2. 3D Visualization of Spatial Data Process
There are a lot of methods about the image of the 3D
visualization, but the basic steps are roughly the same, as
A. TIN G ET AL.
Copyright © 2013 SciRes. OJAppS
shown in Figure 2. First of all, we will input the
two-dimensional images to computer, which convenient
processing format. Through a two-dimensional filter, we
can reduce the image noise effects, improving the signal
to noise ratio and the elimination of image's wake. Take
the medicine image interpolation method, the key parts
of isotropic processing, access to data [1]. After three fil-
ters, different tissues and organs require segmentation
and classification, on the same site of the different image
registration and fusion, in order to facilitate further to a
region of interest of the operation. According to the dif-
ferent requirements of the 3D visualization and system
capacity, choose a different method for 3D rendering, 3D
rec onstr ucti o n.
Fig 2 the medical image 3D visualization steps
2.2. Image visualization algorit hm
Thr ee -dimensional visualization technology is usually
divided into surface rendering and volume rendering of
two methods for.
1) Surface rendering
The method of surface rendering is that, first structure
out of the middle of geometric primitives by the 3D data
(such as surface, plane etc.) and then by the traditional
computer graphics techniques such as reasonable illumi-
nation model, texture mapping method to achieve the
picture drawing [2]. Surface rendering is the most famous
of the marching cubes algorithm (Marching, Cubes,
MC).
2) Volume rendering
Volume rendering is the method to process the volume
data field in each element, namely the voxel. Voxel is the
basic unit in three dimensional, whi c h is composed of
two pieces of adjacent sections of each of the four points
of a c ube [3]. They are generally considered to have cer-
tain properties (opacity and brightness etc), through the
calculation of all voxels on the light, at the same time
interactively adjust opacity, light effects and other para-
meters, and then synthesized with three-dimensional ef-
fect of the image, and direct the 3-D gray-scale data
projection display onto a two-dimensional screen, so
the calculation a large quantity, image generation is
slower, more by hardware technology development
limited, more difficult to implement than the surface
rendering.
The difference of Volume rendering and the tradition
of surface rendering are shown in figure 3. From the im-
age quality, volume rendering is better than surface ren-
der ing. But from the interactive performance and com-
putational efficiency, at least in the current hardware
platform, surface rendering is superior to volume render-
ing, because the surface rendering is the traditional
graphics rendering algorithm, the existing interactive
algorithm and graphics hardware and graphics technolo-
gy can fully play a role [4].
Fig3. The comparison of two kinds of drawing method
3. Reconstruction of 3D Visualization
System's Realization
3.1. VTK's Basic function
The main function of VTK is for image processing,
computer graphics and visualization in scientific compu-
ting, especially for three-dimensional reconstruction of
functions (such as surface rendering and volume render-
ing ability and powerful image processing capability) [5].
VTK graphics and visualization in the field of commonly
used algorithm package into a class, users in visual de-
velopment process does not need to consider the specific
details, which can bring great convenience into research
and development.
The process of VTK visualization is us e the pipelining
mechanism to achieve. That is to say in the visualization
process, above result as an input of next process. Frame
structure as shown in figure 4.
CT/MRI medical
image data
Image processing:
Format conversion
Two dimensional
filtering
Image interpola-
tion
Three -dimensional
filtering
Segmentation
technology
Registration and
fusion
Rendering algo-
rithm
Through the software
tools
Geometry data
Through the sampling
equipmen t
Sampled data set
3D data field
Geometric modeling
Display image
33
A. TING ET AL.
Copyright © 2013 SciRes. OJAppS
Fig 4 the VTK object model framework
First of all, we can establish a data line (Data pipeline)
to process data and link up the Sources Filters and Map-
pers; secondly, establish the appropriate target graphics
to show data.
The system used VTK as the basic architecture and
design the data flow pipeline mode. This design uses B/S
structure, in addition to install some necessary plug-in,
does not need to be installed in the client of other com-
plex software. The system includes the following mod-
ules: the role management module, including hospitals,
doctors, patients, experts, the system administrator; data
module, including the role of books management as well
as the original medical image is processed and data
management; data visualization module, mainly DICOM
Web data visualization and processing results of Web
three -dimensional visualization[6].
The scalability must be considered in module design
fu lly. Several modules must independent design and code:
DICOM data read module, tissue segmentation module,
cutting management module, virtual endoscope mirror
module, 3D reconstruction module, pseudo color man-
agement module, management module, three view sec-
tion frame management module[7].
3D reconstruction in the whole system is at the core
position in management. Module to call the relationship
between such as shown in Figure 5 below:
3.2. System's Realization
1) Development environment
The system is development language for the VC++, in
which use the visualization toolkit VTK.
2) The technology of virtual endoscope
Reconstruction these two-d imensional tomographic
images which obtained through CT and MRI, can show
Fig 5 3D reconstruction data flow diagram
the interior of the body structure [8]. Then the virtual
reality and virtual endoscopic technique, the doctor
thro ugh the mouse to perform various operations, such
as: translational, rotational, parameter setting, in a
three -dimensional internal cavity roaming, simulation
of the traditional endoscope a variety of behavior. Fig-
ure 6 is the virtual endoscope display of human kidney
organ three-dimensional images (a, B, C, D).
a. Input image b. The image is moved and turned
c. The image is cut d. The image is cut and turned
Fig 6. the virtual endoscope display 3D images of human
kidne y
1) Navigation and interaction
The existing navigation mainly has the following sev-
eral methods: manual navigation; users adjust to the
camera directly; through a mouse sliding lens user can
see the inner surface anything, which give user the most
freedom. Guidance navigation that can accord user de-
fined path to adjust the camera and convenient effect
reproduction. It is drawn in a path specific scene, gives
an intuitive representation. Automatic navigation, which
Source
Filter
Mapper
Actor
Render
Render Windows
Properly
Camera Light
Render window Inte-
raction
Data input
VtkImageData
Tissue segmentation
Ps e udo color management
Crop manager set
volume rendering
mapper
Choose the way of
mapping
Implementation
of the corres-
ponding type of
Output
VtkActor
34
A. TIN G ET AL.
Copyright © 2013 SciRes. OJAppS
automatic path generation, automatically generates a na-
vigation path; mark the inner wall of the pipe center line
and fast show pat hs [9].
In this s yst e m, it mainly realized in front of the two
navigation way. For the third navigation way, we need to
study the removal of impurities, multiple branch path
smooth, key technology, which is a very promising me-
thod of navigation and be researching in further.
After loading, the CT pictures displayed on the com-
puter screen, as shown in figure 7.a and figure 7.b.Can be
found by clicking the "lens" button to move the image,
from different angles to observe the structure, whether
the incidence of lesions. Click the "cutting" b utto ns, the
image cutting, and then through the magnification of the
image were observed, helping to make the correct diag-
nosis.
Fig 7. a system interface Fig 7.b CT image loading.
4. Conclusion
These articles realize two-dimensional images of
three -dimensional visualization system wh i c h set out
from the 3D visualization system construction method
and module management two respects. The realization of
the system int e grat ed the information of many aspects. It
is the combination of design about the virtual reality
technology and modern medicine [10]. And it is the mul t i-
disciplinary research. The system is capable of using
multiple imaging or multiple imaging devices for infor-
mation, to provide the defects which caused by the data
information is not accurate or unc er tai nt y [10].
It can be more comprehensive and accurate in clinical
diagnosis, treatment, radiotherapy, plan design, surgical
operation and curative effect evaluation.
REFERENCES
[1] M Haubner, C Krapichler, A Losch, et al. Virtual reality in
medicine- computer graphics and interaction tech-
niques[J]. IEEE Trans.On Information Techology in
Biomedicine, 1997, 1(1):61-72.
[2] Hamann, B.A data reduction scheme for triangulated sur-
fac e[J]. Computer Aided Geometric Design,
1994,11(2):197-214.
[3] Craig Upson, Michael Keeler, V-buffer. Visible volume
rendering [J]. ACM SIGGRAPH Computer Graphics,
1988, 22(4): 59-64.
[4] J Bloomenthal. Polygonalization of Implicit Surfaces[J].
CAGD, 5(4):341-355, 1988.
[5] Ghrare S E,Ali M A M,Jumari K,et.An efficient low com-
plexity lossless coding algorithm for medical im-
ages[J].American Journal of Applied
Sciences,2009,6(8):1502-1508
[6] Kelly, P.J. (1994). Quantitative Virtual Reality Surgical
Simulation, Minimally Invasive Stereotactic Neurosur-
gery and Frameless Stereotactic Technologies. Medicine
Meets Virtual Reality II: Interactive Technology &
Healthcare: Visionary Applications for Simulation Visua-
lization Robotics (pp. 103-108).
[7] Selvaraj K.Data extraction from computer acquired images
of a given 3D environment for enhanced computer vision
and its applications inkinematic design of robs[J].Journal
of Computer Science,2010,6(4):425-427.
[8] Purschke F,et al.Virtual Reality:New Methods for Im-
proving and Accelerating the Development Process in
Vehicle Styling and Design.Computer Graphics Interna-
tional,Proceedings,1998.789.
[9] Yang Mengzhou,Shi Jiaoying.3D Model Reconstruction
Algorithm Based on the Texture of Photos[J].Ruan Jian
Xue Bao/Journal of Software,2000,11(4):502-506.
[10] Xu B, et al. Video Metrics and Optical Methods for 3D
Shape Measurement,Proceedings of SPIE[J].the Interna-
tional Society for Optical Engineering,2001,4309:26-33.
35