Diagnosis of skin cancer using image processing

Esperanza Guerra-Rosas; Josué Álvarez-Borrego; Ángel Coronel-Beltrán

doi:10.1063/1.4897704

Diagnosis of skin cancer using image processing

Esperanza Guerra-Rosas, Josué Álvarez-Borrego^*, Ángel Coronel-Beltrán

^*Corresponding author for this work

Departamento de Investigación en Física

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

4 Scopus citations

Abstract

In this papera methodology for classifying skin cancerin images of dermatologie spots based on spectral analysis using the K-law Fourier non-lineartechnique is presented. The image is segmented and binarized to build the function that contains the interest area. The image is divided into their respective RGB channels to obtain the spectral properties of each channel. The green channel contains more information and therefore this channel is always chosen. This information is point to point multiplied by a binary mask and to this result a Fourier transform is applied written in nonlinear form. If the real part of this spectrum is positive, the spectral density takeunit values, otherwise are zero. Finally the ratio of the sum of the unit values of the spectral density with the sum of values of the binary mask are calculated. This ratio is called spectral index. When the value calculated is in the spectral index range three types of cancer can be detected. Values found out of this range are benign injure.

Original language	English
Title of host publication	International Conference of Computational Methods in Sciences and Engineering 2014, ICCMSE 2014
Editors	Theodore E. Simos, Theodore E. Simos, Theodore E. Simos, Theodore E. Simos, Theodore E. Simos, Zacharoula Kalogiratou, Theodore Monovasilis
Publisher	American Institute of Physics Inc.
Pages	155-158
Number of pages	4
ISBN (Electronic)	9780735412552
DOIs	https://doi.org/10.1063/1.4897704
State	Published - 6 Oct 2014
Event	International Conference of Computational Methods in Sciences and Engineering 2014, ICCMSE 2014 - Athens, Greece Duration: 4 Apr 2014 → 7 Apr 2014

Publication series

Name	AIP Conference Proceedings
Volume	1618
ISSN (Print)	0094-243X
ISSN (Electronic)	1551-7616

Conference

Conference	International Conference of Computational Methods in Sciences and Engineering 2014, ICCMSE 2014
Country/Territory	Greece
City	Athens
Period	4/04/14 → 7/04/14

Bibliographical note

Publisher Copyright:
© 2014 AIP Publishing LLC.

Keywords

Skin cancer
image processing techniques
spectral index

Access to Document

10.1063/1.4897704

Cite this

Guerra-Rosas, E., Álvarez-Borrego, J., & Coronel-Beltrán, Á. (2014). Diagnosis of skin cancer using image processing. In T. E. Simos, T. E. Simos, T. E. Simos, T. E. Simos, T. E. Simos, Z. Kalogiratou, & T. Monovasilis (Eds.), International Conference of Computational Methods in Sciences and Engineering 2014, ICCMSE 2014 (pp. 155-158). (AIP Conference Proceedings; Vol. 1618). American Institute of Physics Inc.. https://doi.org/10.1063/1.4897704

Guerra-Rosas, Esperanza ; Álvarez-Borrego, Josué ; Coronel-Beltrán, Ángel. / Diagnosis of skin cancer using image processing. International Conference of Computational Methods in Sciences and Engineering 2014, ICCMSE 2014. editor / Theodore E. Simos ; Theodore E. Simos ; Theodore E. Simos ; Theodore E. Simos ; Theodore E. Simos ; Zacharoula Kalogiratou ; Theodore Monovasilis. American Institute of Physics Inc., 2014. pp. 155-158 (AIP Conference Proceedings).

@inproceedings{0bdfb302e3054f11b7fd88d36605110a,

title = "Diagnosis of skin cancer using image processing",

abstract = "In this papera methodology for classifying skin cancerin images of dermatologie spots based on spectral analysis using the K-law Fourier non-lineartechnique is presented. The image is segmented and binarized to build the function that contains the interest area. The image is divided into their respective RGB channels to obtain the spectral properties of each channel. The green channel contains more information and therefore this channel is always chosen. This information is point to point multiplied by a binary mask and to this result a Fourier transform is applied written in nonlinear form. If the real part of this spectrum is positive, the spectral density takeunit values, otherwise are zero. Finally the ratio of the sum of the unit values of the spectral density with the sum of values of the binary mask are calculated. This ratio is called spectral index. When the value calculated is in the spectral index range three types of cancer can be detected. Values found out of this range are benign injure.",

keywords = "Skin cancer, image processing techniques, spectral index",

author = "Esperanza Guerra-Rosas and Josu{\'e} {\'A}lvarez-Borrego and {\'A}ngel Coronel-Beltr{\'a}n",

note = "Publisher Copyright: {\textcopyright} 2014 AIP Publishing LLC.; International Conference of Computational Methods in Sciences and Engineering 2014, ICCMSE 2014 ; Conference date: 04-04-2014 Through 07-04-2014",

year = "2014",

month = oct,

day = "6",

doi = "10.1063/1.4897704",

language = "Ingl{\'e}s",

series = "AIP Conference Proceedings",

publisher = "American Institute of Physics Inc.",

pages = "155--158",

editor = "Simos, {Theodore E.} and Simos, {Theodore E.} and Simos, {Theodore E.} and Simos, {Theodore E.} and Simos, {Theodore E.} and Zacharoula Kalogiratou and Theodore Monovasilis",

booktitle = "International Conference of Computational Methods in Sciences and Engineering 2014, ICCMSE 2014",

}

Guerra-Rosas, E, Álvarez-Borrego, J & Coronel-Beltrán, Á 2014, Diagnosis of skin cancer using image processing. in TE Simos, TE Simos, TE Simos, TE Simos, TE Simos, Z Kalogiratou & T Monovasilis (eds), International Conference of Computational Methods in Sciences and Engineering 2014, ICCMSE 2014. AIP Conference Proceedings, vol. 1618, American Institute of Physics Inc., pp. 155-158, International Conference of Computational Methods in Sciences and Engineering 2014, ICCMSE 2014, Athens, Greece, 4/04/14. https://doi.org/10.1063/1.4897704

Diagnosis of skin cancer using image processing. / Guerra-Rosas, Esperanza; Álvarez-Borrego, Josué; Coronel-Beltrán, Ángel.
International Conference of Computational Methods in Sciences and Engineering 2014, ICCMSE 2014. ed. / Theodore E. Simos; Theodore E. Simos; Theodore E. Simos; Theodore E. Simos; Theodore E. Simos; Zacharoula Kalogiratou; Theodore Monovasilis. American Institute of Physics Inc., 2014. p. 155-158 (AIP Conference Proceedings; Vol. 1618).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Diagnosis of skin cancer using image processing

AU - Guerra-Rosas, Esperanza

AU - Álvarez-Borrego, Josué

AU - Coronel-Beltrán, Ángel

PY - 2014/10/6

Y1 - 2014/10/6

N2 - In this papera methodology for classifying skin cancerin images of dermatologie spots based on spectral analysis using the K-law Fourier non-lineartechnique is presented. The image is segmented and binarized to build the function that contains the interest area. The image is divided into their respective RGB channels to obtain the spectral properties of each channel. The green channel contains more information and therefore this channel is always chosen. This information is point to point multiplied by a binary mask and to this result a Fourier transform is applied written in nonlinear form. If the real part of this spectrum is positive, the spectral density takeunit values, otherwise are zero. Finally the ratio of the sum of the unit values of the spectral density with the sum of values of the binary mask are calculated. This ratio is called spectral index. When the value calculated is in the spectral index range three types of cancer can be detected. Values found out of this range are benign injure.

AB - In this papera methodology for classifying skin cancerin images of dermatologie spots based on spectral analysis using the K-law Fourier non-lineartechnique is presented. The image is segmented and binarized to build the function that contains the interest area. The image is divided into their respective RGB channels to obtain the spectral properties of each channel. The green channel contains more information and therefore this channel is always chosen. This information is point to point multiplied by a binary mask and to this result a Fourier transform is applied written in nonlinear form. If the real part of this spectrum is positive, the spectral density takeunit values, otherwise are zero. Finally the ratio of the sum of the unit values of the spectral density with the sum of values of the binary mask are calculated. This ratio is called spectral index. When the value calculated is in the spectral index range three types of cancer can be detected. Values found out of this range are benign injure.

KW - Skin cancer

KW - image processing techniques

KW - spectral index

UR - http://www.scopus.com/inward/record.url?scp=84947544492&partnerID=8YFLogxK

U2 - 10.1063/1.4897704

DO - 10.1063/1.4897704

M3 - Contribución a la conferencia

AN - SCOPUS:84947544492

T3 - AIP Conference Proceedings

SP - 155

EP - 158

BT - International Conference of Computational Methods in Sciences and Engineering 2014, ICCMSE 2014

A2 - Simos, Theodore E.

A2 - Kalogiratou, Zacharoula

A2 - Monovasilis, Theodore

PB - American Institute of Physics Inc.

T2 - International Conference of Computational Methods in Sciences and Engineering 2014, ICCMSE 2014

Y2 - 4 April 2014 through 7 April 2014

ER -

Guerra-Rosas E, Álvarez-Borrego J, Coronel-Beltrán Á. Diagnosis of skin cancer using image processing. In Simos TE, Simos TE, Simos TE, Simos TE, Simos TE, Kalogiratou Z, Monovasilis T, editors, International Conference of Computational Methods in Sciences and Engineering 2014, ICCMSE 2014. American Institute of Physics Inc. 2014. p. 155-158. (AIP Conference Proceedings). doi: 10.1063/1.4897704

Diagnosis of skin cancer using image processing

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this