Abstract – MRI and Mammogram is one of the best [621435]

Abstract – MRI and Mammogram is one of the best
technologies currently being used for diagnosing breast
cancer and brain tumour. Breast cancer and brain tumour is
diagnosed at advanced stages with the help of the mammogram and MRI image. In this thesis an intellige nt
system is designed to diagnose tumour through mammograms, using image processing techniques along with intelligent optimization tools, such as Fire Fly Algorithm (FFA),
Enhanced BEE Colony Optimization (EBCO) and Artificial
Neural Network. The detection of tumour is performed in two
phases: preprocessing and segmentation in the first phase and
feature extraction, selection and classification in the second
phase. 350 MRI images obtained from KMCH Hospital
Coimbatore and 161 pairs of digitized mammograms o btained
from the Mammography Image Analysis Society (MIAS)
database is used to design the proposed diagnosing system.
Initially, the film artifacts and X -ray labels are removed from
the images and median filter is applied to remove the high
frequency components from the image. The suspicious region
is segmented using Markov Random Field (MRF) hybrid with
EBCO and FFA algorithm for MRI and mammogram images. The MRF and EBCO and FFA algorithm based image
segmentation method is a process seeking the optimal l abeling
of the pixels. The optimum label is that which minimizes the
Maximizing a Posterior (MAP) estimate. EBCO and FFA
metaheuristic algorithm is implemented to compute the optimum label, which is to be treated as an optimum threshold for segmentation.
Keywords – MRI, Mammogram , Enhancement, Feature
Extraction, Receiver Operating Characteristics

I. INTRODUCTION
HE segmentation of an image entails the division or
separation of the image into regions of similar attribute.
The ultimate aim in a large number of image processing
applications is to extract important features from the image
data, from which a description, interpretation, or
understanding of the scene can be provided by the machine.
The segmentation of brain tumor from Magnetic Resonance
Images is an important but time -consuming task performed by
medical experts.
Mammography amd MRI images are used to detect tumor
formation within breast and brain tissues. A new nature
A. Sivaramakrishnan, Assistant Professor, Department of Computer
Applications, Karunya University, India. E -mail: [anonimizat]
Dr.M. Karnan, Professor and Head, Department of Computer Science
and Engineering, Tamilnadu College of Engineering, India inspired metaheuristic algorithms remains untouched. The
suspicious region or tumors are segmented using Markov
Random Field hybrid with Enchanced Artificial Bee Colony
Algorithm (EABC) and Fire Fly Algorithm (FFA) (Sahoo,
A. ; Chandra 2013) for mammogram and MRI image
(Karaboga.D, Bahriye Akay 2009).
Firefly algorithm proposed by Yang is a new intelligent
optimization algorithm developed in recent years, Firefly
algorithm is considered as an unconventional swarm -based
heuristic algorithm for constrained optimization tasks inspired
by the flashing behavior of fireflies (Du Xiaogang et.al 2013).
Fire Fly Algorithm (FFA) is a recent population -based
approach inspired by the observation of real firefly and based upon their brightness behaviour. In FFA, solutions of the problem are constructed within an iterative process, by adding
solution components to partial solutions. Each individual
FireFly constructs a part of the solution using a brightness and distance, which reflects its experience accumulated while
solving the problem, and heuristic information dependent on
the problem.
Recently, many researchers have focused their attention on
a new class of Algorithms called metaheuristics. A
metaheuristic is a set of algorithmic concepts that can be used
to define heuristic methods applicable to a wide set of different problems. In other words, a metaheuristic can be seen as a
general -purpose heuristic method designed to guide an
underlying problem specific heuristic toward promising
regions of the search space containing high- quality solutions.
A metaheuristic therefore a general algorithmic framework,
which can be applied to different optimization problems with relatively few modifications to make them, adapted to a
specific problem. The use of metaheuristics has significantly
increased the ability of finding very high -quality solutions to
hard, practically relevant combinatorial optimization problems
in a reasonable time. This is particularly true for large and
poorly understood problems. Several meta -heuristics, such as
GA, Ant Colony Optimization (ACO), PSO, Tabu Search and
Simulated Annealing, have been proposed to deal with the
computationally intractable problems. FFA is a new meta –
heuristic developed for composing approximate solutions.
A detailed study on methods of various stages of automatic
detection of microcalcification and brain tumours in d igital
mammogram and MRIs. It is to be noted that researchers have not used Fire Fly Algorithm to analyse the mammogram and
MRI in the recent past. In this work, the metaheuristic
algorithms such as FFA and EABC are implemented to extract the suspicious re gion based on texture image segmentation. Medical Image Segmentation Using Firefly
Algorithm and Enhanced Bee Colony Optimization
A. Sivaramakrishnan and Dr. M. Karnan
T
International Conference on Information and Image Processing (ICIIP-2014)
316
ISBN 978-93-83459-16-2 © 2014 Bonfring

In the mammogram and MRI image segmentation process,
a pioneering method, viz., Markov Random Field hybrid with
Enchanced Artificial Bee colony algorithm and Fire Fly
Algorithm is used to segment the tumour from t he
mammogram and MRI image.
The MRF based image segmentation method is a process
of seeking the optimal labeling of the pixels. The optimum label is that which minimizes the Maximizing a Posterior estimate. I nitially, a unique label is assigned for simila r
patterns to the mammogram and MRI images. The Enchanced
Artificial Bee colony algorithm and Fire Fly Algorithm
algorithm is applied to obtain the optimum label, which is to
be considered an optimum threshold for segmentation. Figure1 shows the Flow Diagrams for the Medical Image segmentation.
II. M
ARKOV RANDOM FILED
The image is stored in a two- dimensional matrix and a
kernel is extracted for each pixel. A unique label is assigned to the kernels having similar patterns. In the labeling process, a label matri x is initialized with zeros. The size of the label
matrix is equal to the size of the image. For each pixel in the image, the label value is stored in the label matrix at the location corresponding to its central pixel coordinates in the
gray level image

Fig. 1 : Flow Diagrams for the Medical Image Segmentation
A pattern matrix is maintained to store the dissimilar
patterns in the image. For each pixel, a kernel is extracted and the kernel is c ompared with the patterns available in the
pattern matrix. Once it finds any matches the same label value
is assigned to the currently extracted kernel. Otherwise the
next label value is assigned to the kernel and the kernel is
added to the pattern matrix. The labels are assigned integer
values starting with one and incremented by one whenever a new pattern occurs. Finally the pattern matrix contains all the dissimilar patterns in the image and the corresponding label
values are also extracted from the label matrix.
The challenge of finding the MAP estimate of the
segmentation is to search for the optimum label which
minimizes the posterior energy function U(x). In this section a
new effective approach, Enchanced Artificial Bee colony
algorithm and Fire Fly Algorithm is proposed for the
minimization of MAP estimation.
III. S
EGMENTATION OF MAMMOGRAM AND MRI IMAGE
USING FIRE FLY ALGORITHM (FFA)
The Firefly algorithm was developed by Xin- She Yang, on
the basis of flashing light behavior of fireflies in nature ( Ming –
Huwi Horng ; Ting -WeiJiang 2010 ; Goutam Das2013) .
A. Fire Fly Metaheuristic
The fundamental function of the glowing light is to attract
mate during mating session, where male firefly uses a brief signal pattern and female firefly respond in certain time interval for the same species. The pattern of flashes is often
unique for a particular species of fireflies. The brightness of a
firefly is affected or determined by the landscape of the fitness value to be optimized. For a maximization problem, the
brightness is simply proportional to the value of the fitness
value. The Figure 2 shows the FFA Algorithms
The general algorithm of firefly:
Define the fitness function of f(x), where x=(x 1… x d) T
Generate the initial population of fireflies or x i (i=1, 2,.., n)
Determine the light intensity of Ii at x i via f(x i)
While (t<MaxGen)
For i = 1 to n (all n fireflies);
For j=1 to n (n fireflies)
if (Ij > Ii), move firefly i towards j;
end if
Attractiveness varies with distance r via Exp [ -γr2];
Evaluate new solutions and update light intensity;
End for j;
End for i;
Rank the fireflies and find the current best;
End while;
Postprocess results and visualisation;
End procedure;
Fig. 2: FFA Algorithms
Alpha determines random percentage in firefly moving. It
includes value between zeros to one. Absorption coefficient is
named gamma. The constraint varies between zeros to
extreme. If Coefficient is close to zero, then ß=ß0 and this corresponds to a special case of particle swarm optimization.
Besides, if absorption coefficient is close particularly, this is
the case where the fireflies fly in a very foggy region randomly. Finally ß0 is maximum Attractiveness value. Th e
algorithm constitutes a population -based iterative procedure
with numerous agents (perceived as fire flies) concurrently
solving a considered optimization problem. Agents MRF to obtain MAP values
MRF to obtain MAP values
EABC to obtain
optimum threshold
with FCM values

Fire Fly Algorithm
(FFA) to obtain
optimum threshold
Segmented Image FCM Enhanced Mammogram
Image
International Conference on Information and Image Processing (ICIIP-2014)
317
ISBN 978-93-83459-16-2 © 2014 Bonfring

communicate with each other via bioluminescent glowing
which enables them to explor e cost function space more
effectively than in standard distributed random search.
Intelligence optimization technique is based on the
assumption that solution of an optimization problem can be perceived as agent (fire fly) which glows proportionally to i ts
quality in a considered problem setting. Consequently each brighter fire fly attracts its partners (regardless of their sex), which makes the search space being explored more efficiently.
The firefly algorithm has three particular idealized rules which
are based on some of the basic flashing characteristics of real
fireflies. They are the following:
• All fireflies are unisex and they will move towards
more attractive and brighter ones regardless of their sex.
• The degree of attractiveness of a firefly is proportional
to its brightness. Also the brightness may decrease as the distance from the other fire flies increases due to
the fact that the air absorbs light. If there is not a
brighter or more attractive fire fly than a particular one it will then move randomly. The brightness or light
intensity of a fire fly is determined by the value of the objective function of a given problem. . Based on these three rules, the basic steps of the Fire Fly Algorithm
(FFA) can be summarized as the pseudo code shown
in Fig.4.3.
Objective function f (x), x = ( x1, …, x d)T
Generate initial population of fireflies xi ( i = 1, 2, …, n )
Light intensity Ii at xi is determined by f(xi)
Define light absorption coefficient
while ( t <MaxGeneration)
for i = 1 : n all n fireflies
for j = 1 : i all n fireflies
if (Ij > Ii), Move firefly i towards j in d-dimension; end if
Attractiveness varies with distance r via exp[− r]
Evaluate new solutions and update light intensity
end for j; end for i
Rank the fireflies and find the current best
end while
Postprocess results and visualization
Figure 3: Pseudo Code of the Fire Fly Algorithm
B. Attractiveness
In the firefly algorithm, there are two important issues: the
variation of light intensity and formulation of the
attractiveness. Assume that the attractiveness of a firefly is
determined by its brightness which in turn is associated with the encoded objective function. In the simplest case for
maximum optimization problems, the brightness If a firefly at
a particular location x can be chosen as I(x) / f(x).
However, the attractiveness is relative, it should be seen in
the eyes of the beholder or judged by the other fireflies. Thus,
it will vary with the distance rij between firefly i and firefly j.
In addition, light intensity decreases with the distance from its
source, and light is also absorbed in the media, so we should allow the attractiveness to vary with the degree of absorption. In the simplest form, the light intensity I(r) varies
according to the inverse square law I(r ) = Is/r2 where I s is the
intensity at the source. For a given medium with a fixed light absorption coefficient , the light intensity I varies with the
distance r. That is I = I0e−r, where I0 is the original light
intensity. In order to avoid the singularity at r = 0 in the
expression Is/r2, the combined effect of both tinverse square
law and absorption can be approximated using the Gaussian
form.
C. Distance and Movement
Euclidean distance
In the Euclidean plane , if p = ( p1, p2) and q = ( q1, q2) then
the distance is given by

This is equivalent to the Pythagorean theorem .
Alternatively, it follows from ( 2) that if the polar
coordinates of the point p are (r1, θ1) and those of q are (r2, θ2),
then the distance between the points is

The distance between any two fireflies i and j at xi and xj,
respectively, is the
Cartesian distance
rij = ||xi − xj || = vuutdX
k=1 (xi, k − xj, k)2, where xi,k is the kth component of the
spatial coordinate xi of ith firefly. In2 -D image, we have r ij =
q(xi − xj)2 + (yi − yj)2.
The movement of a firefly i is attracted to another more
attractive (brighter) firefly j is determined byxi = xi + 0 e−r2: ij
(xj − xi) + _ (rand −12), where the second term is due to the
attraction while the third term is random – ization with being
the randomization parameter. rand is a random number
generator uniformly distributed in [0, 1].
D. Extracting Suspicious Region from Image using Fire Fly
Algorithm
Fire Fly Algorithm is applied to find the optimum label
from the pattern matrix. Initially, the dissimilar patterns, the
corresponding labels and the MAP values are stored in a
solution matrix and the parameters such as number of iterations ( NI), number of firefly (NF), γ – the light absorption
coefficient r: the particular distance from the light source d: the domain space are assigned the values of 1, 10 and 0.001 respectively. Also the solution matrix contains separate
columns for absorpt ion, distance and flag values of each
firefly.
The flag value is used to indicate whether the kernel has
been selected previously or not. Initially all the flag values are set to zero and the absorption, distance values are assigned d0.
At the initial step , all the fireflies are assigned random kernels
and the pheromone values are updated. The posterior energy
function value for all the selected kernels from each firefly is
extracted from the solution matrix. Compare the posterior
energy function value for all the selected kernels from each
International Conference on Information and Image Processing (ICIIP-2014)
318
ISBN 978-93-83459-16-2 © 2014 Bonfring

firefly, to select the minimum value from the set, which is
known as ‘Local Minimum’ (Lmin) or ‘Iterations best’
solution. This local minimum value is again compared with the
‘Global Minimum’ (Gmin). If the local minimum is less than the global minimum, then the local minimum is assigned with
the current global minimum. Then the kernel that generates
this local minimum value is selected and its brightness is updated.
The brightness value for the remaining kernels is upda ted.
Thus the brightness and distance values are updated globally.
This procedure is repeated for all the image pixels. At the final
iteration, the Gmin has the optimum label of the image. The
corresponding kernel is selected from the pattern matrix. The intensity value of the center pixel in the kernel is selected as
optimum threshold value for segmentation. In the MRI image, the pixels having lower intensity values than the threshold value are changed to zero. The entire procedure is repeated for
any numb er of times to obtain the more approximated value.
E. Fire Fly Algorithm (FFA) With FCM
After completing all the process by Fire Fly Algorithm the
generated output is given to the FCM as input. The optimal value of FFA through mammogram or MRI Brain Image is
given as an input for FCM. The aim of FCM is to find cluster
centers (centroids) that minimize dissimilarity function.
In the first step, the algorithm selects the initial cluster
from FFA Algorithm. Then, in later step after several iteration
of the algorithm, the final result converges to actual cluster of
FFA with FCM. The Maximum Adaptive threshold is used to
compare the current neuron value,If the current value is less than the Adaptive Thresholds neglects the region set to black
and the suspicious region is look like bright. The aim of FFA
with FCM is to detect the suspicious region from the
background region in the mammogram or MRI brain Image.
Input: mammogram or MRI Brain Image
Output: Segmented Image contains only Tumor
(suspicious reign)
Step 1. Read the brain image or the stored in a two
dimensional matrix
Step 2. Divide the image to 3×3 sub image (cells)
Step 3. For each label in the image, calculate the post erior
energy U (x) value
U(x) ={Σ[( y-μ)2
/(2*σ2
)]+Σ log(σ)+Σ V(x)}
where
y = intensity value of Pixels in the kernel,
μ = mean value of the kernel,
σ = standard deviation of the kernel,
V = potential function of the kernel, and
x = center Pixel of the label. If x1 is equal to x2 in a kernel,
then
V(x) = β, otherwise 0, where β is visibility relative
parameter (β≥0)
Step 4. The posterior energy values of all the labels are
stored in a separate matrix
Step 5. FFA is used to mini mize the posterior energy function.
The procedure is as follows:
Step 6. Initialize the values of number of iterations (N),
number of fireflies (K),
Step 7. Create a solution matrix (S) to store the labels of all
the Pixels, posterior energy values of all the
Pixels, initial brightness values for all the fireflies at each
pixels , and a flag column to mention whether the pixels is
selected by the firefly or not
Step 8. Store the labels and the energy function values in S
Step 9. Initialize the light obsorbtion coefficient and
distance values,
Step 10. Initialize all the flag values for the entire firefly
with 0, it means that pixels is not selected yet, if it is set to 1
means selected
Step 11. Select a random pixel for each Ant, which is not
selected previously
Step 12. Update the pheromone values for the selected
pixels by all the firefly
Step 13. Select the minimum value from the set, assign as
local minimum(Lmin)
Step 14. Compare this loca l minimum (Lmin) with the
global minimum (Gmin), if Lmin is less than Gmin, assign
Gmin=Lmin
Step 15. Select the Ant, whose solution is equal to local
minimum, to update its Brightness and distance globally
Step 16. Perform the steps (13) to (15) till all the Image
Pixels have been selected
Step 17. Perform the steps (7) to (16) for M times
Step 18. The Gmin has the optimum label which minimizes
the posterior energy function
Step 19. Gmin (Global optimal value) has the optimum is
taken from center (3×3) value of optimal label
Step 20. Give a center cluster value is a Gmin
In the image, the pixels having lower intensity values than
the threshold value are changed to zero. The entire
procedure is repeated for any number of times to obtain the
more approximated value.
In the mammogram and MRI image segmentation process,
a swarm intelligence method, viz., FFA and EABC is used to segment the tumour from the mammogram and MRI image .
The figure 4.6 shows the segmented Mammogram image using

Figure s: Segmented Mammogram Image using the FF
Algorithm
International Conference on Information and Image Processing (ICIIP-2014)
319
ISBN 978-93-83459-16-2 © 2014 Bonfring

Figure: Segmented MRI Image using the FF Algorithm
IV. SEGMENTATION OF MAMMOGRAM AND MRI IMAGE
USING HYBRID MRF – ENHANCED ARTIFICIAL BEE
COLONY
Image segmentation has been approached from a wide
variety of perspectives such as region -based approach,
morphological operation, multi -scale analysis and fuzzy
approaches and stochastic approaches are discussed earlier. In
this section, MRF hybrid with Enchanced Artificial Bee
Colony (EABC) is implemented for mammogram and MRI
image segmentation.
A. Enchanced Artificial Bee Colony
Enchanced Artificial Bee Colony (EABC) algorithm is a
new swarm intelligence method which simulates intelligent foraging behavior of honey bees. Artificial Bee Colony (ABC)
system is a n ovel optimization algorithm inspired of the
natural behavior of honey bees in their search process for the
best food sources, which proposed by Karaboga and Basturk in
2006. ABC consists of three groups of bees: employed ,
onlooker and scout bees (Karaboga , D, Basturk, .B. 2006,
2007)
After identifying the initial population and the nectar value,
the employed bee and onlooker bee can be applied to generate the new population. Employed bee operation produces a new
string for onlooker bee operation. Stochasti c selection process
is implemented as linear search through roulette wheel with
slots weighted in proportion to kernel fitness values. In this
function, a random number multiplies the sum of the
population fitness called as the stopping point.
B. Algorithm: Segmentation using Enhanced Artificial Bee
Colony
Oij ← Original Image;
Xij ← segmented image;
[m n] ← size of O ij
for each pixel in O ij
Gi← Population ←intensity of the Original pixel from O ij,
converted to binary string, i ←1 to n
F← Nectar (or )fitness value; computed by MAP
Pop1 ← initial population contains G
End
(Employed bee Operation – production mechanism)
p ← 1
repeat for N times (N=300)
for each string in Population1
g1, g2 ← select two strings for stochastic selection proc ess.
(Onlooker bee Operation) For the selection of two strings roulette wheel is implemented
as follows:r = random() * sum_of_Nectar or fitness
[Hint: random() function returns a random number between 0
and 1]
(i) Fsum = 0, i=0, m → size of the population
(ii) F → contains the population strings
Fsum = Fsum +F(i) → nectar value 0 to 1
i = i + 1
if ( (i <= m) and (Fsum < r)) Goto Step: (iv)
return i
g3, g4 ←sharing the valuable information.
Pop2(p) ← g3;
p ← p+1; Pop2(p) ← g4;
end
(Scout bee Operation)
min ← Min(Pop2); Pop1 ← Pop2;
pos ← where the X(i,j) = min;
end
Fig: Algorithm for Segmentation using MRF -Enchanced
Artificial Bee Colony Algorithm
The Figure: 10 shows the Algorithm of EABC with FCM
In the mammogram and MRI image segmentation process, a
pione ering method, viz., Markov Random Field hybrid with
Artificial Bee Colony Optimization algorithm is used to segment the microcalcifications from the mammogram and MRI image . Figure 4.11 shows the segmented image using the
EABC algorithm

Figure: Segmented Mammogram Image Using the EABC
algorithm

Figure: Segmented MRI Image using the EABC Algorithm
V. EXPERIMENTS AND RESULTS
The effectiveness of the proposed technique is determined
by extracting the suspicious region from the mammogram and MRI image using FFA and EABC. The true positive detection
rate and the number of false positive detection rate at various
thresholds of the segmented images are used to measure the
algorithm’s performance. These rates are represented using Receiver Operating C haracteristic curves.
International Conference on Information and Image Processing (ICIIP-2014)
320
ISBN 978-93-83459-16-2 © 2014 Bonfring

True Positive (TP) and False Positive (FP) rates are
calculated at ten different thresholds selected on the segmented
image to generate the ROC curve. Previous methods are taken
an overlap region of only 40% as true positive. However, in
the proposed method, the true positive is considered only at 80%
of overlap. All other regions extracted by the algorithm are
labeled false positives. Figure 13 shows the ROC curves generated on the full test set, using 10 operating points from
FFA and EABC segmentation. If the threshold value is low
true detections may become merged with false positive regions

Fig: Combined results on all 161 normal and abnormal MIAS
image pairs using FUZZY, FFA and MRF -EABC
Classification Ratio: The area under the ROC curve (Az
value) is an importFFA criterion for evaluating diagnostic performance. The ROC curve is in the range between zero and
one. The value of Az is 1.0 when the diagnostic detection has perfect performance, which means that TP rate is 100% and FP
rate is 0%. The Az value for the proposed algorithm is 0.99. Table 1 shows the comparison of detection rate between the previous works and the proposed method.
Table 1 : Performance Analysis
Sl.No Author Method Result
The tumor
detection rate
(Az value)
1 S. Murugavalli and
V. Rajamani 2007 FUZZY 93.21
2 T. Logeswari and
M.Karnan2010 HSOM 93.21
3
The Proposed
approach
EABC 96.00
FFA 99.13
VI. CONCLUSION
The three different methods such as FFA and ABC
segmentation techniques for mammogram and MRI image
segmentation have been implemented. The suspicious region
was extracted from the mammogram and MRI image based on the combination of Markov Random Field wit h FFA and
Enhanced Artificial Bee Colony Optimization. In MRF the
image pixels are labelled and their posterior function values
were computed. FFA and EABC were used to find the
optimum label that minimizes the Maximizing a Posterior
estimate to segment th e image. To evaluate the performance of
the segmentation algorithms the ROC curve was generated.
The experimental results show that the FFA withfuzzy
produces 0.99 and the EABC withfuzzy method produces 0.96 as Az value. It was observed that the metaheuris tic FFA hybrid
with MRF performed well.
R
EFERENCES
[1] Karaboga.D, Bahriye Akay,” A comparative study of Artificial Bee Colony algorithm” Journal of Applied Mathematics and Computation, Elsevier Inc, 214 (2009) 108– 132. 2009
[2] Karaboga, D, Basturk,.B, A powerful and efficient algorithm for numerical function optimization: artificial bee colony (abc) algorithm, Journal of Global Optimization 39 (3) (2007) 459– 471.2007
[3] Karaboga,.D, Basturk.B, On the performance of artificial bee colony (abc) algorithm, Applied Soft Computing 8 (1) (2008) 687– 697.2008
[4] Du Xiaogang ; Dang Jianwu ; Wang Yangping ; Liu Xinguo ; Li
Sha “An Algorithm Multi- Resolution Medical Image Registration Based
on Firefly AlgorithmanPowell”
Intelligent System Design and Engineering Applications (ISDEA), 2013
Third International Conference on Publication Year: 2013 , Page(s): 274
– 277 2013
[5] Sahoo, A. ; Chandra, S. L'evy -flight firefly algorithm based active
contour model for medical image segmentation 2013 Sixth International
Conference on Contemporary Computing , Page(s): 159 – 162 2013
[6] Ming -Huwi Horng ; Ting -WeiJiang 2010 ; Goutam Das2013 Multilevel
Image Thresholding Selection Based on the Firefly Algorithm
Autonomic & Trusted Computing (UIC/ATC), 2010 7th Inter national
Conference on Ubiquitous Intelligence & Computing 2010 , Page(s): 58
– 63 2010
[7] Goutam Das, Md. Iqbal Quraishi, Manisha Barman firefly algorithm
based image analysis Asian journal of computer science & information technology vol 3, no 11 (2013)
[8] T. Logeswari and M.Karnan, "An Improved Implementation of Brain Tumor Detection Using Segmentation Based on Hierarchical Self Organizing Map", International Journal of Computer Theory and
Engineering, Vol. 2, No. 4, August, 2010.
[9] S. Murugavalli and V. Rajamani An Improved Implementation of Brain
Tumor Detection Using Segmentation Based on Neuro Fuzzy
International Conference on Information and Image Processing (ICIIP-2014)
321
ISBN 978-93-83459-16-2 © 2014 Bonfring