Journal of Basic and Applied Engineering Research
Print ISSN: 2350-0077; Online ISSN: 2350-0255; Volume 1, Number 6; October, 2014 pp. 22-26
© Krishi Sanskriti Publications
http://www.krishisanskriti.org/jbaer.html
Noise Removal from Brain Image
using Region Filling Technique
Daizy Deb
1
, Sudipta Roy
2
1,2
Dept. of Information Technology, Assam University, Silchar, Assam, India
Abstract: In today’s world, one of the reason in rise of mortality
among the people is brain cancer. Brain tumour is the main cause
of brain cancer. A tumour can be defined as any mass caused by
abnormal or uncontrolled growth of cells. This mass of tumour
grows within the skull, due to which normal brain activity is
hampered. Which is if not detected in earlier stage, can take away
the person’s life. Hence, it is very important to detect the brain
tumour as early as possible. For detection of brain tumour, first we
have to read the MRI image of brain and then we can apply
segmentation on the image. But in the MRI brain image, some
confidential information of patient’s is always there. To apply
segmentation, this unnecessary information has to be removed, as it
can be considered as noise. Here we present an efficient method for
removing noise from the MRI image of brain using Region filling
method.
Keywords: Brain tumor, Noise, Filtering, Region of Interest,
Region Filling.
1. INTRODUCTION
Brain cancer is one of the leading causes of death in the world
now days. An uncontrolled growth of cancer cells in the brain
leads to brain cancer, which is a very serious type of
malignancy. A malignant brain tumour is the main cause of
brain cancer. All brain tumours are not malignant, some are
benign also. Brain cancer is also called glioma and
meningioma [1].
According to the National Brain Tumour Society, US, over
600,000 people are living with the primary brain tumour.
Among these 600,000 people, 28,000 are children under the
age of 20. Metastatic brain tumours (cancer that spreads from
other parts of the body to the brain) are the most common type
of brain tumour, which is the reason of cancer for 20% to 40%
of persons. Over 7% of all the primary brain tumours reported
in the United States are diagnosed among children under the
age of 20. 210,000 people in the United States are diagnosed
with a primary or metastatic brain tumour every year i.e. over
575 people a day.
In general, the risk of developing a malignant CNS or brain
tumour over the course of one’s lifetime is less than 1%. But
the risk increases with the age. 4.5 per 100,000 persons under
the age of 20 will be diagnosed with a malignant brain tumour.
After the age of 75, this rate rises to 57 per 100,000 persons.
Among the people over the age of85, the risk stops increasing.
The risk for developing brain cancer is very high among the
people with a family history of brain cancer and those who
had radiation therapy of the head.
2. RELATED WORKS IN NOISE REMOVAL
T. Logeswari and M. Karnan [1] applied weighted median
filter for removing the noise presented in the MRI image of
the brain. Weighted median filter is a type of nonlinear filters.
It retains the robustness and edge preserving capacity of the
image. Dr. Samir Kumar Bandyopadhyay [3] removed noise
based on Maximum Difference Threshold value, which is
constant threshold value determined by observation. Pratibha
Sharma and co-authors [4] applied spatial noise filter for
removing noise from the MRI image of a brain. Sudipta Roy,
Samir K. Bandyopadhyay [5] first used high pass filter and
then finally used median filter for removing noise. Here a high
pass filter is used in matlab, by which each pixel of the image
is replaced by weighted average of the surrounding pixels.
Then merging of gray scale image and filtered image is done
for enhancing the image quality. Median filter is applied to the
enhanced image. High pass filter is used by Rajesh C. Patil,
Dr. A. S. Bhalchandra [6] for removing noise and then they
applied median filter to enhance the quality of the image.
Noise removal has to be done in such a way that it should not
affect the portion of the brain in the image as each portion is
the most important part to detect the tumor. Hence noise
removal should not blur the image.
3. NOISE AND MEDICAL IMAGE
“Noise” originally means “unwanted signal” i.e. noise
represents unwanted information which deteriorates image
quality. The process which affects the acquired image and is
not part of the scene can be defined as noise. A random
variation of brightness or color information in images can also
be termed as noise [10].