A method to detect potentially malignant skin lesions through image segmentation

Carlos A. Wilches; Óscar J. Perdomo; César A. Perdomo

doi:10.1007/978-981-10-9035-6_52

A method to detect potentially malignant skin lesions through image segmentation

Carlos A. Wilches, Óscar J. Perdomo, César A. Perdomo

Resultado de la investigación: Contribución a una revista › Artículo de la conferencia › revisión exhaustiva

Resumen

Melanoma is a form of skin cancer responsible for most cancer deaths. The detection of melanoma in an early stage is still a challenge with high importance. The aim of this paper was the design of a system to analyse the mole’s characteristics, based on the ABCDE criteria, for identifying potentially malignant skin lesions. The segmentation was based on Otzu thresholding, chain code and skeletonization methods were used to compute the main features: asymmetry, border, color and texture, for a future classification of skin lesions as normal or potentially malignant melanoma. The system was evaluated successfully using 92 images with 27 benign and 65 malignant melanomas from Dermatology Information System (DermIS).

Idioma original	Inglés estadounidense
Páginas (desde-hasta)	289-293
Número de páginas	5
Publicación	IFMBE Proceedings
Volumen	68
N.º	1
DOI	https://doi.org/10.1007/978-981-10-9035-6_52
Estado	Publicada - 2019
Evento	World Congress on Medical Physics and Biomedical Engineering, WC 2018 - Prague, República Checa Duración: jun 3 2018 → jun 8 2018

All Science Journal Classification (ASJC) codes

Bioingeniería
Ingeniería biomédica

ODS de las Naciones Unidas

Este resultado contribuye a los siguientes Objetivos de Desarrollo Sostenible

Acceder al documento

10.1007/978-981-10-9035-6_52

Otros archivos y enlaces

Citar esto

@article{b9731171a0d244288d5e720810f9613d,

title = "A method to detect potentially malignant skin lesions through image segmentation",

abstract = "Melanoma is a form of skin cancer responsible for most cancer deaths. The detection of melanoma in an early stage is still a challenge with high importance. The aim of this paper was the design of a system to analyse the mole{\textquoteright}s characteristics, based on the ABCDE criteria, for identifying potentially malignant skin lesions. The segmentation was based on Otzu thresholding, chain code and skeletonization methods were used to compute the main features: asymmetry, border, color and texture, for a future classification of skin lesions as normal or potentially malignant melanoma. The system was evaluated successfully using 92 images with 27 benign and 65 malignant melanomas from Dermatology Information System (DermIS).",

author = "Wilches, {Carlos A.} and Perdomo, {{\'O}scar J.} and Perdomo, {C{\'e}sar A.}",

note = "Publisher Copyright: {\textcopyright} Springer Nature Singapore Pte Ltd. 2019. Copyright: Copyright 2018 Elsevier B.V., All rights reserved.; World Congress on Medical Physics and Biomedical Engineering, WC 2018 ; Conference date: 03-06-2018 Through 08-06-2018",

year = "2019",

doi = "10.1007/978-981-10-9035-6_52",

language = "English (US)",

volume = "68",

pages = "289--293",

journal = "IFMBE Proceedings",

issn = "1680-0737",

publisher = "Springer",

number = "1",

}

TY - JOUR

T1 - A method to detect potentially malignant skin lesions through image segmentation

AU - Wilches, Carlos A.

AU - Perdomo, Óscar J.

AU - Perdomo, César A.

PY - 2019

Y1 - 2019

N2 - Melanoma is a form of skin cancer responsible for most cancer deaths. The detection of melanoma in an early stage is still a challenge with high importance. The aim of this paper was the design of a system to analyse the mole’s characteristics, based on the ABCDE criteria, for identifying potentially malignant skin lesions. The segmentation was based on Otzu thresholding, chain code and skeletonization methods were used to compute the main features: asymmetry, border, color and texture, for a future classification of skin lesions as normal or potentially malignant melanoma. The system was evaluated successfully using 92 images with 27 benign and 65 malignant melanomas from Dermatology Information System (DermIS).

AB - Melanoma is a form of skin cancer responsible for most cancer deaths. The detection of melanoma in an early stage is still a challenge with high importance. The aim of this paper was the design of a system to analyse the mole’s characteristics, based on the ABCDE criteria, for identifying potentially malignant skin lesions. The segmentation was based on Otzu thresholding, chain code and skeletonization methods were used to compute the main features: asymmetry, border, color and texture, for a future classification of skin lesions as normal or potentially malignant melanoma. The system was evaluated successfully using 92 images with 27 benign and 65 malignant melanomas from Dermatology Information System (DermIS).

UR - http://www.scopus.com/inward/record.url?scp=85048266565&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85048266565&partnerID=8YFLogxK

U2 - 10.1007/978-981-10-9035-6_52

DO - 10.1007/978-981-10-9035-6_52

M3 - Conference article

AN - SCOPUS:85048266565

VL - 68

SP - 289

EP - 293

JO - IFMBE Proceedings

JF - IFMBE Proceedings

SN - 1680-0737

IS - 1

T2 - World Congress on Medical Physics and Biomedical Engineering, WC 2018

Y2 - 3 June 2018 through 8 June 2018

ER -

A method to detect potentially malignant skin lesions through image segmentation

Resumen

All Science Journal Classification (ASJC) codes

ODS de las Naciones Unidas

Acceder al documento

Otros archivos y enlaces

Huella

Citar esto