IRJET- Three-Dimensional Analysis on Dermoscopic Images with RSA Encrypted Diagnosis

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 06 Issue: 04 | Apr 2019 www.irjet.net p-ISSN: 2395-0072
© 2019, IRJET | Impact Factor value: 7.211 | ISO 9001:2008 Certified Journal | Page 547
THREE-DIMENSIONAL ANALYSIS ON DERMOSCOPIC IMAGES WITH RSA
ENCRYPTED DIAGNOSIS
Divya Jayanadhan Nair1, Vishnu Prabha N Kaimal2
1M-Techstudent, Applied Electronics and Communication Engineering, NCERC,Kerala,India
2Assistant professor, Dept of Electronics and Communication Engineering, NCERC,Kerala,India
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - There are many skin diseases in the present
world. This leads to the death of n number of human beings.
The main objective of the proposed work is to detect the early
stages of melanoma. This proposed system introduces a
noninvasive computerized dermoscopy system with RSA
encryption .This methodmainlyconcentrateinthe detectionof
skin cancer depth and the result of analysis is secured by RSA
encryption method. The skin lesion images of the patient
which is in the 2D form are converted into 3 D image after the
depth estimation.Thus,the 3D skin Lesion image isobtainedby
the depth analysis of the images. A 3D skin lesion
reconstruction technique using the estimated depth obtained
from regular dermoscopic images is presented. On basis of the
3D reconstruction, depth and 3D shape featuresareextracted.
In addition to 3D designed to diagnose basal cell carcinoma,
blue nevus, dermatofibroma, haemangioma, seborrhoeic
keratosis and normal mole lesions. For experimental
evaluations ISIC 2016: Melanoma Project datasets is
considered which is freely available on net. Significant
performance improvement is reported post inclusion of
estimated depth and 3D features. RSA encryption will be
applied on the result generated after the diagnosis.
Key Words: 3D skin lesion, melanoma, blue nevus,
dermatofibroma, haemangioma, seborrhoeic keratosis,
ISIC dataset, RSA encryption.
1. INTRODUCTION
As there are the increasing skin diseases in the world,
the world health organizationdeclaresthatskincancer
is the increasing disease [1]. Thus, there are various
types of skin cancer can be classified as melanoma and
the non-melanoma category [2].The diagnosis of the
skin disease can be done at the early stages. The death
rate due to the skin cancer in the united states alone
will be 75%.The person who are having the melanoma
skin cancer will die more than the person having non-
melanoma skin cancer[3][4].There is an annual
increase in the death rate per year is 2.6%.Thus,the
early detection of skin cancer melanoma is achieved.
The five year survival rate of 95% is reported. The
survival rate of skin cancer is 13% [5]. Thus, the early
detection of skin cancer is the important factor. The
proposed method consists of dermoscopy technique
which consists of non-invasivetechnique.Dermoscopy
is the technique which is performed as applyingthegel
liquid in the lesion region. This method is known as
stereomicroscope or dermatoscope which is used to
magnifying the images. They are characterized by the
skin color, structure and the pattern data of the skin
region. The data which is gathered cannot be visible
with the human eye. Thus, the skin lesions are
identified with the above data and diagnosis them
[7].Many dermatologists use many algorithm such as
ABCD (Asymmetry, Border, Color and Diameter) [8],
ABCDE (Asymmetry, Border, Color, Diameter and
Evolution) [9]

Menzies method [10] and the seven-point checklist
[11].These method are undergoes by the diagnosis of
skin cancer .By these method, detection of skin cancer
is about 5-30% easy and accurate. This will be easy for
the practice of dermatologists.Thesecannotbecarried
out using naked eye [12].
2. LITERATURE SURVEY
A Multi Parameter Extraction and Classification
System(MPECS) is proposed to distinguish early
melanoma[12].A six stage approach is[11]adopted to
remove the colour,texture and shape highlights.
Grouping of three skin injury types, to be specific
"Progressed Melanoma","Non-Melanoma,"Early
Melanoma" is accomplished and not clarifiedaboutthe
precise profundity of the tumor into favorable or
harmful sorts. The utilization of traditional clinical
calculations, for example, ABCD (Asymmetry, Border,
Color and Diameter) [8], ABCDE (Asymmetry, Border,
Color, Diameter and Evolution) [9],Menzies strategy
[10] and the seven-point agenda [11] is receivedbyfor
the analysis of melanoma skin sores.
3. METHODOLOGY
The image processing techniques consists of following
steps.
3.1 IMAGE ACQUISITION
Image acquisition is getting the input image through
various sources such as camera or through x-ray, MRI
scanning, CT scan and microscopic analysis. Thus, the
image acquisition can be done in various input format
such as jpeg, jpg, bmp and png format.
3.2 PRE-PROCESSING
Pre-processing stages consist of various methods and
various steps included inimageprocessingtechniques.
They include
3.3 GRAY-SCALE CONVERSION
Preprocessing stage consists of color conversion and
filtering process. The input retinal images can beinthe
form of RGB images. These RGB images are converted
into gray scale images.RGB images are converted into
gray scale images due to the elimination of hue and
saturation of the input images.
3.4 FILTERING
Filtering process is done to eliminate the noise within
the image. The filter which is used in the proposed
work is the median filter. Median filter is used to
eliminate the noise within the retinal images and also
used to smoothen the retinal images. By eliminating
the salt and pepper noise, the skin lesionimagescanbe
smoothen.
3.5 SEGMENTATION
Segmentation process is carried out to separate the
lesion region from the other regions of the skin. The
segmented part undergoes the various classification
procedures to find the early stages of skin lesions.
3.6 CLASSIFICATION
The classification process can be undergoes for the
early stage detection of skin lesions in the human.
There are various stages of classifiers which is namely
SVM(support Vector Machine),Multi class SVM and
ANN (Artificial Neural Networks).They are used to

analysis the various stages of skin cancer in the
diseased persons.
4. PROPOSED SYSTEM
The Block diagram of the proposed workcanbeshown
in the figure 4.1
Figure 4.1 Block diagram of the proposed work.
4.1 IMAGE ACQUISITION
The input 3 dimensional images are acquired form the
clinical database. They have the collection of various
skin lesions patient images. More than n number of
patient’s stages can be analyzed using image
processing technique. The image acquisition stage is
followed by the pre-processing stage.
4.2 PRE-PROCESSING
This stage consists of various steps such as gray-scale
image, filtering process. Thus, the output of these
stages are shown in the figure 4.2
Figure 4.2 Graphical user interface short screen.
4.3 SEGMENTATION
Pre-processing stages is followed by the segmentation
process. The segmentation process is carriedoutusing
adaptive snake model algorithm .This algorithm is
carried out to separate the minor regions of the skin
lesions .The final stages of segmentation in shown in
the figure 4.3
Figure 4.3 Segmentation
Mapping is done to segment the lesion region and to
find the feature extracted from the segmented region.
Thus, the intermediate result is shown in the figure4.4

Figure 4.4 Intermediate result of Feature extraction
4.4 FEATURE EXTRACTION
Feature extraction is considered as the importantstep,
it helps in the classification process .In this paper
3Dreconstruction of image is done from the
2Ddermoscopic image by estimating the depth. On the
basis of 3D reconstruction ,3D shape feature,2D shape
feature ,color and texture are extracted, in figure 4.6
final 3D reconstruction of image is shown from
2Ddermoscopic image.
4.5 CLASSIFICATION
Final step is the classification process, the accuracy of
classification mainly depends on how accurately
segmentation and feature extraction is done the
classification process is carried out using Two
classifiers used here Ada boost and multi SVM to
compare the performance of classifier. This will
identify the type of skin lesions.
Figure 4.5 Blur scale graph representation
Figure 4.6 Final 3D reconstruction of image from 2D
dermoscopy image.
Two classifiers used here Ada boost and multi SVM to
compare the performance of classifier.
Figure 4.7 Final classification

4.5 PERFORMANCE METRICS
Thus, the accuracy and the encryption and decryption
process is carried out as shown in the figure 4.8 and in
figure 4.9 performance graph of two classifiers is
shown.
Figure 4.8 Accuracy.
Figure 4.9 short screen of performance graph.
From the performance graph of two classifiers ada
boost and Multi-Svm ,it is clear that the accuracy rate
Multi Svm is more(98.69%) than the Ado boost
classifier (92%).Final result is encrypted for security
purpose, RSA algorithm is used, and 32 bit key is used
here.
Figure 4.10 Encrypted data.
Figure 4.11 Decrypted data.
5. CONCLUSION
Thus, by the proposed method there is a reduction in
the time consumption. It will be useful for the practice
of the dermatologist. IN future, classification process
may change with better algorithm which would
analysis more number of diseases.
REFERENCES
[1]WHO inter sun (2015)
http://www.who.int/uv/faq/skincancer/en/index1.ht
ml (accessed 21 July 2015.). [2] L. Baldwin and J. Dunn,
“Global Controversies and Advances in Skin Cancer,”
Asian Pacific Journal of Cancer Prevention, vol. 14, no.
4, pp. 2155-2157, 2013.
[3] American Cancer Society, Cancer Facts & Figures
2015: American Cancer Society 2015
[4] N. Howlader, A. Noone, M. Krapcho, J. Garshell, N.
Neyman,S. Altekruse, C. Kosary, M. Yu, J. Ruhl, Z.
Tatalovich, H. Cho,A. Mariotto, D. Lewis, H. Chen, E.
Feuer, and K. Cronin. (2012,Apr.). “SEER cancer
statistics review, 1975-2010,” National Cancer
Institute,Bethesda, MD, USA
[5] U.S. Emerging Melanoma Therapeutics Market,
a090-52, Tech. Rep.,2001

[6] H. Pehamberger, A. Steiner, and K. Wolff, “In vivo
epiluminescence microscopy of pigmented skin
lesions—I: Pattern analysis of pigmentedskinlesions,”
J. Amer. Acad. Dermatol., vol. 17, pp. 571– 583, 1987.
[7] J. Mayer, “Systematic review of the diagnostic
accuracy of dermatoscopy in detecting malignant
melanoma,” Med. J. Aust., vol. 167, no. 4, pp. 206–210,
Aug. 1997.
[8] W. Stolz, A. Riemann, and A. Cognetta,“ABCDruleof
dermatoscopy: A new practical method for early
recognition of malignant melanoma,” Eur. J. Dermatol.,
vol. 4, pp. 521–527, 1994.
[9] A. Blum, G. Rassner, and C. Garbe, “Modified abc-
point list of dermoscopy:A simplified and highly
accurate dermoscopic algorithm for the diagnosis of
cutaneous melanocytic lesions,” J. Am Acad.
Dermatol.,vol. 48, no. 5, pp. 672–678, May 2003.
[10] S. Menzies, C. Ingvar, K. Crotty, and W. McCarthy,
“Frequencyandmorphologiccharacteristicsofinvasive
melanomas lacking specific surface microscopic
features,” Arch. Dermatol., vol. 132, no. 10, pp. 1178–
1182, Oct. 1996.
[11] G. Argenziano, G. Fabbrocini, P. Carli, V. De Giorgi,
E. Sammarco, and M. Delfino, “Epiluminescence
microscopy for the diagnosis of doubtful melanocytic
skin lesion.

IRJET- Three-Dimensional Analysis on Dermoscopic Images with RSA Encrypted Diagnosis

More Related Content

What's hot (20)

Similar to IRJET- Three-Dimensional Analysis on Dermoscopic Images with RSA Encrypted Diagnosis (20)

More from IRJET Journal (20)

Recently uploaded (20)

IRJET- Three-Dimensional Analysis on Dermoscopic Images with RSA Encrypted Diagnosis