[attachment=13575]
FPGA BASED DESIGN AND VERIFICATION OF IMAGE SEGMENTATION FOR MEDICAL SYSTEMS AND DATA PROCESSING
AIM OF THE PROJECT
Performing the medical image segmentation using FPGA hardware kit.
Image Segmentation is to divide an image into different regions to understand clearly about the error.
In our project image segmentation can be done by using clustering technique.
For segmenting the segmentation process, we need 3 softwares.
Matlab,
EDK tool
visual basic 6.0
MEDICAL IMAGE SEGMENTATION
Medical image is nothing but gray image i.e.,
result of CT Scan or MRI scan.
Image segmentation in medical images is performing the clustering technique on 4 clusters image.
Medical image segmentation is to provide richer information than in ordinary medical image
The desired output contains the labels that classify the input gray scale voxel.
The collection of labels that produced under segmentation is called “labelmap”.
The segmentation and gray scale data can provide by the Dr. Martha Shenton’s Schizophrenia Research Group.
Applications of medical image segmentation
3D visualization : Segmentatioin allows creation of 3D surface models.
Advantage of surface model is that it gives a
3D view of any image and is improvement over 2D image view.
Volumetric Measurement : Measurement of the volumes of anatomical structures is necessary in medical studies, both of normal anatomy and of various pathological conditions or disorders.
For example, in studies of schizophrenia, volume measurement is used to quantify the variation in neural anatomy between schizophrenic and control patients.
Volume measurement is also used to diagnose patients; one example is in measurement of the ejection fraction.
BLOCK DIAGRAM
MODIFIED BLOCK DIAGRAM
GSC ALGORITHM
GSC: Gray value Structure Code used to convert 3D into 2D image
In Medical field, image is represent in 3D space
The 3 dimensional specification indicates location value and density
This 3D space medical image contains small volume elements
Those volume elements called voxel
voxel = volume + pixel
In GSC we have two method
1. Linking phase
2. Splitting phase
Linking phase :
By linking the nearest neighbouring voxel, it can be rejoined to form segmented image.
Splitting phase:
The formed element is then divided into small voxels.
K-MEANS CLUSTERING METHOD FOR SEGMENTATION
CLUSTERING TECHNIQUE
TOOLS & SOFTWARE
MATLAB
Read
a=imread (‘input.bmp’);
write
b=imwrite (a,’xyz.bmp’);
To display image
imshow(‘a’);
pixel conversion
GUI
GUI: Graphical user interface to display the input image
CREATING A GUI IN MATLAB
DIALOG BOX
FPGA KIT AND EDK TOOLS
It provides the unique features of FPGA family and provides a convinient development board for applications
DDR SDRAM
SLIDE SWITCHES
DISCRETS LEDs
UART CABLE
USB TYPE-A/TYPE-B CONNECTORS
XPS AND SDK
USB Type B connector
Interfacing the FPGA kit to user computer
Xilinx kit
Internal components
SDRAM
The kit includes a standard USB Type A/Type B cable
The wider and narrower Type A connector fits the USB connector at the back of the computer
square Type B connector to the Spartan-3E Starter Kit board
The Spartan-3E Starter Kit board has four slide switches
The slide switches are located in the lower right corner of the board and are labeled SW3 through SW0.
Switch SW3 is the left-most switch, and SW0 is the right-most switch.