Skin Cancer Detection

N. Durga Rao et al. Int. Journal of Engineering Research and Application www.ijera.com
ISSN : 2248-9622, Vol. 6, Issue 6, ( Part -4) June 2016, pp.60-63
www.ijera.com 60|P a g e
Skin Cancer Detection
N. Durga Rao*
, Dr.G.Sudhavani#
*Asst Professor, Vnitsw, Guntur, Andhrapradesh, India
#Professor R.V.R & J.C, Guntur, Andhrapradesh, India
ABSTRACT: In recent days, skin cancer is seen as one of the most Hazardous form of the Cancers found in
Humans. Skin cancer is found in various types such as Melanoma, Basal and Squamous cell Carcinoma among
which Melanoma is the most unpredictable. The detection of Melanoma cancer in early stage can be helpful to
cure it. Computer vision can play important role in Medical Image Diagnosis and it has been proved by many
existing systems. In this paper, we present a survey on different steps which are being to detect the Melanoma
Skin Cancer using Image Processing tools. In every step, what are the different methods are be included in our
paper.
Keywords: ABCD Rule, Melanoma, preprocessing, segmentation.
I. INTRODUCTION
Cancer begins when cells in the part
of the body starts to grow out of control. A
lesion means possibly abnormal change or
difference in a tissue or structure such as the skin.
Cancer is an uncontrolled growth of abnormal
cells. The skin cancer is the uncontrolled growth
of skin cells in the body. It develops when an
unrepaired DNA damage to the skin cells and
mostly caused by ultraviolet radiation of the sun
or tanning beds, trigger mutations (genetic
defects) which leads to the skin cells multiply
rapidly and malignant tumors form. Some skin
cancer can spread and cause damage in the near
by tissue cells [1]. Also, in some cases, skin
cancer can be on vital organs. Sun is the most
common cause Of skin cancer. But it fully does not
explain that skin cancer usually develop on the skin
exposed to sunlight. Also it can be exposed to
environmental threats, radiation analysis, and
even inheritance could play a role. Although
anyone can get skin cancer, the risk is greatest
for people having bright skin or bright eyes, a
wealth of large, irregular shaped moles, a genetic
family history of skin cancer, an excessive
sunlight or sunburn, lived in large or with year
round sunshine, received radiation medication [2].
The sign of skin cancer often starts as the
change of color in the skin. They are usually
mixed color (pink, red, and brown). There are
three types of skin cancer that occurred. They are-
Basel cell cancer, squamous cell carcinoma and
malignant melanoma tumor. The first two does not
spread quickly, but the third one spreads quickly.
Melanoma is much less common than basal cell
and squamous cell skin cancer, but it is far more
dangerous than the other two types. However, it is
much more dangerous if it is not found early. It
causes the majority (75%) of deaths related to
the skin cancer [3]. Worldwide, doctors
investigate about 160,000 new cases of melanoma
per year. It is mostly common in women than in
men. In women, the most common part that the
cancer occurs is the legs and in case of men
the most common part is on the back. .It is
specifically common among Caucasians, and
especially north Europeans those who lives in
sunny climates. There are high rates of
occurrences in Australia, North and South
America, and North Europe, also with a
disordered decrease in southern Italy and Sicily.
II. SKIN CANCER DETECTION
At first an image is acquired with a
digital camera under consistent lighting. The
proper interpretation of these dermoscopic
images leads to increased clinical diagnostic
accuracy. Most Automated Skin Lesion Diagnosis
methods adopt the standard computer-aided
diagnosis (CAD) pipeline which is illustrated in
Fig: 1 and it consists of five general stages. After
the image is acquired, it contains many artifacts
such as hair and oil bubbles which could bias
downstream processes are identified. Next, the
lesion is segmented from the surrounding healthy
skin. After segmentation, discriminative features
are extracted from the lesion. Features which are
usually extracted are border, color, entropy,
compactness, radial variance of the mask,
coarseness. Finally; by extracting these features
the detection is done which finally shows the risk
probability of the lesion which is present in the
image.
2.1 Preprocessing
In this section, we describe the
preprocessing steps that facilitate the border
detection procedure namely, (1) Color space
RESEARCH ARTICLE OPEN ACCESS

transformation (2) Contrast enhancement (3)
Artifact removal.
2.1.1 Color Space Transformation
Dermoscopy images are commonly
acquired using a digital camera with a dermoscope
attachment. Due to the computational simplicity
and convenience of scalar (single channel)
processing, the resulting RGB (red-green-blue)
color image is often converted to a scalar image
using one of the following methods:
(i) Retaining only the blue channel (lesions are
often more prominent in this channel).
(ii) Applying the luminance transformation,
(iii) Applying the Karhunen-Loéve (KL)
transformation [6] and retaining the channel
with the highest variance.
Fig: 1 Flowchart of skin cancer detection
2.1.2 Contrast Enhancement
One of the factors that complicate the
detection of borders in dermoscopy images is
insufficient contrast. Recently, Delgado et al. [4]
proposed a contrast enhancement method based on
Independent Histogram Pursuit (IHP). This
algorithm linearly transforms the original RGB
image to a decorrelated color space in which the
lesion and the background skin are maximally
separated. Border detection is then performed on
these contrast enhanced images using a simple
clustering algorithm.
2.1.3. Artifact Removal
Dermoscopy images often contain artifacts
such as such as black frames, ink markings, rulers,
air bubbles, as well as intrinsic coetaneous features
that can affect border detection such as blood
vessels, hairs, and skin lines. These artifacts and
extraneous elements complicate the border
detection procedure, which results in loss of
accuracy as well as an increase in computational
time. The most straightforward way to remove
these artifacts is to smooth the image using a
general purpose filter such as the Gaussian (GF),
Median (MF), or Anisotropic Diffusion Filters
(ADF). Several issues should be considered while
using these filters:
 Scalar vs. vector processing: These filters are
originally formulated for scalar images. For
vector images one can apply a scalar filter to
each channel independently and then combine
the results, a strategy referred to as marginal
filtering. Although fast, this scheme introduces
color artifacts in the output. An alternative
solution is to use filters that treat the pixels as
vectors [5].
 Mask size: The amount of smoothing is
proportional to the mask size. However,
excessively large masks result in the blurring
of edges, which might reduce the border
detection accuracy. Setting the mask size
proportional to the image size seems to be a
reasonable strategy [6,7].
 Computational time: For the GF and MF,
algorithms that perform in constant time
independent of the mask size have been
developed [8,9]. As for the ADF, the
computational time depends on the mask size
and the number of iterations.
2.2 Segmentation
Image segmentation is the process of
separating or grouping an image into different
parts. These parts normally correspond to
something that humans can easily separate and
view as individual objects. Computers have no
means of intelligently recognizing objects, and so
many different methods have been developed in
order to segment images
2.2.1. Methods for Image Segmentation
Image segmentation methods can be
categorized as below (1) Region Based Methods
(2) Edge Based Methods (3) Hybrid Techniques
2.3Feature Extraction & Detection
Before extracting features from the skin
lesion and classifying the lesion as malignant or
benign, the location of the lesion border must be
identified using a segmentation algorithm. Finding
an accurate estimate of the lesion border is
important because of the types of features used for
classification. One common set of features is the
ABCD scale: asymmetry, border irregularity, color
variegation, and diameter.
To calculate the ABCD score, the
'Asymmetry, Border, Colors, and Dermoscopic
structures' criteria are assessed semi quantitatively
.Each of the criteria is then multiplied by a given
weight factor to yield a Total Dermoscopy Score
(TDS).
TDS values less than 4.75 indicate a
benign melanocytic lesion, values between 4.8 and

5.45 indicate a suspicious lesion, and values of 5.45
or greater are highly suggestive of melanoma.
(1) Asymmetry: To assess asymmetry, the
melanocytic lesion is bisected by two 90º axes
that were positioned to produce the lowest
possible asymmetry score. If both axes
dermocopically show asymmetric contours
with regard to shape, colors and/or
dermoscopic structures, the asymmetry score is
2. If there is asymmetry on one axis only, the
score is 1. If asymmetry is absent with regard
to both axes the score is 0.
(2) Border: The lesion is divided into eighths, and
the pigment pattern is assessed. Within
each one-eighth segment, a sharp, abrupt cut-
off of pigment pattern at the periphery
receives a score 1. In contrast, a gradual,
indistinct cut-off within the segment receives a
score of 0. Thus, the maximum border score is
8 and the minimum score is 0.
(3) Color: Six different colors are counted in
determining the color score: white, red, light
brown, dark brown, blue-gray, and black. For
each color present, add +1 to the score. White
should be counted only if the area is lighter
than the adjacent skin. The maximum color
score is 6 and the minimum score is 1.
(4) Dermoscopic Structures: Evaluation of
dermoscopic structures focuses on 5 structural
features: network, structure less (or
homogeneous) areas, branched streaks, dots,
and globules. The presence of any feature
results in a score +1 Structure less (or
homogenous) areas must be larger than 10% of
the lesion to be considered present. Branched
streaks and dots are counted only when more
than two are clearly visible. The presence of a
single globule is sufficient for the lesion to be
considered positive for globules.
Formula for TDS: [(A score x 1.3) + (B score x
0.1) + (C score x 0.5) + (D score x 0.5)]
Table 1: The ABCD Score & Weight Factor
Table 2: TDS Score Interpretation
Total Dermoscopy
Score (TDS)
Interpretation
<4.75 Benign melanocytic
lesion
4.8-5.45 Suspicious lesion; close
follow-up or excision
recommended
>5.45 Lesion highly
suggestive of melanoma
False-positive
score (>5.45)
sometimes
observed in:
 Reed and Spitz
nevus
 Clark nevus with
globular pattern
 Congenital
melanocytic nevus
2.4. Postprocessing
The result of the segmentation procedure
is either a label image or a binary edge map. In
order to obtain the lesion border, the segmentation
output should be post processed. The precise
sequence of post processing operations depends on
the particular choice of the segmentation method.
However, certain operations seem to be useful in
general. These include:
 Region merging: Ideally, the segmentation
procedure is expected to produce two regions:
the lesion and the background skin. However,
since these regions are rarely homogeneous,
segmentation methods often partition them into
multiple subregions. In order to obtain a single
lesion object, subregions that are part of the
lesion should first be identified and then
merged.
 Border smoothing: Most segmentation
methods produce regions with ragged borders.
More natural borders can be obtained by a
variety of operations including majority
filtering [5], morphological filtering [6], and
curve fitting.
 Border expansion: In several studies, it was
observed that the computer-detected borders
were mostly contained within the

dermatologist-determined borders. This is
because the automated segmentation methods
tend to find the sharpest pigment change,
whereas the dermatologists choose the outmost
detectable pigment. The discrepancy between
the two borders can be reduced by expanding
the computer detected border using
morphological filtering [5], Euclidean distance
transform [5], or iterative region growing [10].
III. CONCLUSION
In this study, we have discussed various
methodologies for the melanoma diagnosis,
combination of image processing and soft
computing techniques yielded more accurate results
to detect melanoma which will act as an aid for
clinical diagnosis. The process of melanoma is
carried out in various stages like preprocessing,
segmentation, feature extraction, post processing
which employ sophisticated techniques for getting
accurate results.
References
[1]. Yana Goncharova, Enas A. S. Attia, Khawla
Souid, and Inna V. Vasilenko,
“Dermoscopic Features of Facial
Pigmented Skin Lesions,” Hindawi
Publishing Corporation, Dermatology, pp. 1-
7, 2013
[2]. Iris Zalaudek, M.D, Aimilios Lallas,
Elvira Moscarella, Caterina Longo, H. Peter
Soyer, Giuseppe Argenziano, “The
dermatologist’s applications traditional and
new applications of dermoscopy,”
Dermatol Pract Concept, vol.3, no. 2, pp.
67-71, April 30, 2013.
[3]. Cheng Lu,Muhammad Mahmood, Naresh
Jha, and Mrinal Mandal, “Automated
Segmentation of the Melanocytes in Skin
Histopathological Images,” IEEE Journal
Of Biomedical And Health Informatics,
Vol. 17, No. 2, Pp. 284-296, March 2013.
[4]. Delgado D, Butakoff C, Ersboll BK,
Stoecker WV. “Independent Histogram
Pursuit for Segmentation of Skin Lesions”
IEEE Trans on Biomedical Engineering
2008; 55(1):157–161.
[5]. Celebi ME, Kingravi HA, Aslandogan YA.
“Nonlinear Vector Filtering for Impulsive
Noise Removal from Color Images”. Journal
of Electronic Imaging 2007; 16(3):033008.
(21 pages)
[6]. Schmid P. “Segmentation of Digitized
Dermatoscopic Images by Two-Dimensional
Color Clustering”. IEEE Trans on Medical
Imaging 1999; 18(2):164–171.
[7]. Celebi ME, Kingravi HA, Iyatomi H, et al.
“Border Detection in Dermoscopy Images
Using Statistical Region Merging. Skin
Research and Technology” 2008; 14(3):347–
353. [Pub Med: 19159382]
[8]. Geusebroek J-M, Smeulders AWM, van de
Weijer J. “Fast Anisotropic Gauss Filtering”.
IEEE Trans on Image Processing 2003;
12(8):938–943.
[9]. Perreault S, Hébert P.” Median Filtering in
Constant Time.” IEEE Trans on Image
Processing 2007; 16 (9):2389–2394.
[10]. Iyatomi H, Oka H, Saito M, et al.
“Quantitative Assessment of Tumor
Extraction from Dermoscopy Images and
Evaluation of Computer-based Extraction
Methods for Automatic Melanoma
Diagnostic System”. Melanoma Research
2006; 16(2):183–190. [Pub Med: 16567974]

Skin Cancer Detection

More Related Content

What's hot

Viewers also liked

Similar to Skin Cancer Detection

Recently uploaded

Skin Cancer Detection