A Deep Convolutional Neural Network for Melanoma Recognition in Dermoscopy Images

Sahar Nasiri Haghighi; Habibollah Danyali; Mohammad Sadegh Helfroush; Mohammad Hasan Karami

doi:10.1109/ICCKE50421.2020.9303684

A Deep Convolutional Neural Network for Melanoma Recognition in Dermoscopy Images

Sahar Nasiri Haghighi
, Habibollah Danyali
, Mohammad Sadegh Helfroush
, Mohammad Hasan Karami

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

4 Citations (Scopus)

Abstract

Automated melanoma recognition in dermoscopy images is a challenging task due to a set of hindrances including low contrast skin images, the resemblance of melanoma and non-melanoma skin lesions, and the great variety in this type of skin cancer. However, in this study, a fully automated method is proposed which recognizes the melanoma lesions from the non-melanoma lesions with high accuracy. Convolutional Neural Networks (CNNs) have made great strides in the field of recognition and classification of medical images. Based on this ground, a deep convolutional neural network is proposed that acts as the central pillar of the proposed melanoma recognition method. In order to compensate for the lack of training data, data augmentation techniques have been employed. The proposed method is a merger of the features elicited from the proposed Convolutional Neural Network architecture and a Support Vector Machine (SVM) classifier. The classifier categorizes the input dermoscopy images into two main classes of Melanoma and non-Melanoma skin lesion images with a promising accuracy of 89.52\%, which outperforms the state-of-art methods.

Original language	English
Title of host publication	2020 10th International Conference on Computer and Knowledge Engineering (ICCKE)
Publisher	Institute of Electrical and Electronics Engineers
Pages	453-456
Number of pages	4
ISBN (Electronic)	978-1-7281-8566-8
ISBN (Print)	978-1-7281-8567-5
DOIs	https://doi.org/10.1109/ICCKE50421.2020.9303684
Publication status	Published - 31 Dec 2020
Externally published	Yes
Event	10th International Conference on Computer and Knowledge Engineering, ICCKE 2020 - Mashhad, Iran, Islamic Republic of Duration: 29 Oct 2020 → 30 Oct 2020 Conference number: 10

Conference

Conference	10th International Conference on Computer and Knowledge Engineering, ICCKE 2020
Abbreviated title	ICCKE 2020
Country/Territory	Iran, Islamic Republic of
City	Mashhad
Period	29/10/20 → 30/10/20

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being

Keywords

Lesions
Melanoma
Skin
Support vector machines
Feature extraction
Image recognition
Convolutional neural networks
data augmentation
convolutional neural networks
super vector machine
Melanoma recognition

Access to Document

10.1109/ICCKE50421.2020.9303684

Cite this

@inproceedings{3e958fdef20545ffb8bf2481e366a9c1,

title = "A Deep Convolutional Neural Network for Melanoma Recognition in Dermoscopy Images",

abstract = "Automated melanoma recognition in dermoscopy images is a challenging task due to a set of hindrances including low contrast skin images, the resemblance of melanoma and non-melanoma skin lesions, and the great variety in this type of skin cancer. However, in this study, a fully automated method is proposed which recognizes the melanoma lesions from the non-melanoma lesions with high accuracy. Convolutional Neural Networks (CNNs) have made great strides in the field of recognition and classification of medical images. Based on this ground, a deep convolutional neural network is proposed that acts as the central pillar of the proposed melanoma recognition method. In order to compensate for the lack of training data, data augmentation techniques have been employed. The proposed method is a merger of the features elicited from the proposed Convolutional Neural Network architecture and a Support Vector Machine (SVM) classifier. The classifier categorizes the input dermoscopy images into two main classes of Melanoma and non-Melanoma skin lesion images with a promising accuracy of 89.52\textbackslash{}\%, which outperforms the state-of-art methods.",

keywords = "Lesions, Melanoma, Skin, Support vector machines, Feature extraction, Image recognition, Convolutional neural networks, data augmentation, convolutional neural networks, super vector machine, Melanoma recognition",

author = "Haghighi, \{Sahar Nasiri\} and Habibollah Danyali and Helfroush, \{Mohammad Sadegh\} and Karami, \{Mohammad Hasan\}",

year = "2020",

month = dec,

day = "31",

doi = "10.1109/ICCKE50421.2020.9303684",

language = "English",

isbn = "978-1-7281-8567-5",

pages = "453--456",

booktitle = "2020 10th International Conference on Computer and Knowledge Engineering (ICCKE)",

publisher = "Institute of Electrical and Electronics Engineers",

address = "United States",

note = "10th International Conference on Computer and Knowledge Engineering, ICCKE 2020, ICCKE 2020 ; Conference date: 29-10-2020 Through 30-10-2020",

}

Haghighi, SN, Danyali, H, Helfroush, MS & Karami, MH 2020, A Deep Convolutional Neural Network for Melanoma Recognition in Dermoscopy Images. in 2020 10th International Conference on Computer and Knowledge Engineering (ICCKE)., 9303684, Institute of Electrical and Electronics Engineers, pp. 453-456, 10th International Conference on Computer and Knowledge Engineering, ICCKE 2020, Mashhad, Iran, Islamic Republic of, 29/10/20. https://doi.org/10.1109/ICCKE50421.2020.9303684

A Deep Convolutional Neural Network for Melanoma Recognition in Dermoscopy Images. / Haghighi, Sahar Nasiri; Danyali, Habibollah; Helfroush, Mohammad Sadegh et al.
2020 10th International Conference on Computer and Knowledge Engineering (ICCKE). Institute of Electrical and Electronics Engineers, 2020. p. 453-456 9303684.

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

TY - GEN

T1 - A Deep Convolutional Neural Network for Melanoma Recognition in Dermoscopy Images

AU - Haghighi, Sahar Nasiri

AU - Danyali, Habibollah

AU - Helfroush, Mohammad Sadegh

AU - Karami, Mohammad Hasan

N1 - Conference code: 10

PY - 2020/12/31

Y1 - 2020/12/31

N2 - Automated melanoma recognition in dermoscopy images is a challenging task due to a set of hindrances including low contrast skin images, the resemblance of melanoma and non-melanoma skin lesions, and the great variety in this type of skin cancer. However, in this study, a fully automated method is proposed which recognizes the melanoma lesions from the non-melanoma lesions with high accuracy. Convolutional Neural Networks (CNNs) have made great strides in the field of recognition and classification of medical images. Based on this ground, a deep convolutional neural network is proposed that acts as the central pillar of the proposed melanoma recognition method. In order to compensate for the lack of training data, data augmentation techniques have been employed. The proposed method is a merger of the features elicited from the proposed Convolutional Neural Network architecture and a Support Vector Machine (SVM) classifier. The classifier categorizes the input dermoscopy images into two main classes of Melanoma and non-Melanoma skin lesion images with a promising accuracy of 89.52\%, which outperforms the state-of-art methods.

AB - Automated melanoma recognition in dermoscopy images is a challenging task due to a set of hindrances including low contrast skin images, the resemblance of melanoma and non-melanoma skin lesions, and the great variety in this type of skin cancer. However, in this study, a fully automated method is proposed which recognizes the melanoma lesions from the non-melanoma lesions with high accuracy. Convolutional Neural Networks (CNNs) have made great strides in the field of recognition and classification of medical images. Based on this ground, a deep convolutional neural network is proposed that acts as the central pillar of the proposed melanoma recognition method. In order to compensate for the lack of training data, data augmentation techniques have been employed. The proposed method is a merger of the features elicited from the proposed Convolutional Neural Network architecture and a Support Vector Machine (SVM) classifier. The classifier categorizes the input dermoscopy images into two main classes of Melanoma and non-Melanoma skin lesion images with a promising accuracy of 89.52\%, which outperforms the state-of-art methods.

KW - Lesions

KW - Melanoma

KW - Skin

KW - Support vector machines

KW - Feature extraction

KW - Image recognition

KW - Convolutional neural networks

KW - data augmentation

KW - convolutional neural networks

KW - super vector machine

KW - Melanoma recognition

UR - https://www.scopus.com/pages/publications/85101447464

U2 - 10.1109/ICCKE50421.2020.9303684

DO - 10.1109/ICCKE50421.2020.9303684

M3 - Conference contribution

SN - 978-1-7281-8567-5

SP - 453

EP - 456

BT - 2020 10th International Conference on Computer and Knowledge Engineering (ICCKE)

PB - Institute of Electrical and Electronics Engineers

T2 - 10th International Conference on Computer and Knowledge Engineering, ICCKE 2020

Y2 - 29 October 2020 through 30 October 2020

ER -

A Deep Convolutional Neural Network for Melanoma Recognition in Dermoscopy Images

Abstract

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this