Two Systems for the Detection of Melanomas in Dermoscopy Images Using Texture and Color Features

被引:252
作者
Barata, Catarina [1 ]
Ruela, Margarida [1 ]
Francisco, Mariana [1 ]
Mendonca, Teresa [2 ]
Marques, Jorge S. [1 ]
机构
[1] Inst Super Tecn, Inst Syst & Robot, P-1049001 Lisbon, Portugal
[2] Univ Porto, Fac Ciencias, P-4169007 Oporto, Portugal
来源
IEEE SYSTEMS JOURNAL | 2014年 / 8卷 / 03期
关键词
Bag of features (BoF); color; computer-aided diagnosis; dermoscopy; melanoma; texture; BORDER DETECTION; EPILUMINESCENCE MICROSCOPY; SKIN-LESIONS; ABCD RULE; CLASSIFICATION; DIAGNOSIS; DERMATOSCOPY; RECOGNITION; EXTRACTION; ALGORITHM;
D O I
10.1109/JSYST.2013.2271540
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Melanoma is one of the deadliest forms of cancer; hence, great effort has been put into the development of diagnosis methods for this disease. This paper addresses two different systems for the detection of melanomas in dermoscopy images. The first system uses global methods to classify skin lesions, whereas the second system uses local features and the bag-of-features classifier. This paper aims at determining the best system for skin lesion classification. The other objective is to compare the role of color and texture features in lesion classification and determine which set of features is more discriminative. It is concluded that color features outperform texture features when used alone and that both methods achieve very good results, i.e., Sensitivity = 96% and Specificity = 80% for global methods against Sensitivity = 100% and Specificity = 75% for local methods. The classification results were obtained on a data set of 176 dermoscopy images from Hospital Pedro Hispano, Matosinhos.
引用
收藏
页码:965 / 979
页数:15
相关论文
共 55 条
[41]   Automated diagnosis of pigmented skin lesions [J].
Rubegni, P ;
Cevenini, G ;
Burroni, M ;
Perotti, R ;
Dell'Eva, G ;
Sbano, P ;
Miracco, C ;
Luzi, P ;
Tosi, P ;
Barbini, P ;
Andreassi, L .
INTERNATIONAL JOURNAL OF CANCER, 2002, 101 (06) :576-580
[42]   A novel method for detection of pigment network in dermoscopic images using graphs [J].
Sadeghi, Maryam ;
Razmara, Majid ;
Lee, Tim K. ;
Atkins, M. Stella .
COMPUTERIZED MEDICAL IMAGING AND GRAPHICS, 2011, 35 (02) :137-143
[43]   Pigment distribution in melanocytic lesion images: a digital parameter to be employed for computer-aided diagnosis [J].
Seidenari, S ;
Pellacani, G ;
Grana, C .
SKIN RESEARCH AND TECHNOLOGY, 2005, 11 (04) :236-241
[44]   Comparison of Segmentation Methods for Melanoma Diagnosis in Dermoscopy Images [J].
Silveira, Margarida ;
Nascimento, Jacinto C. ;
Marques, Jorge S. ;
Marcal, Andre R. S. ;
Mendonca, Teresa ;
Yamauchi, Syogo ;
Maeda, Junji ;
Rozeira, Jorge .
IEEE JOURNAL OF SELECTED TOPICS IN SIGNAL PROCESSING, 2009, 3 (01) :35-45
[45]   Malignant Melanoma Detection by Bag-of-Features Classification [J].
Situ, Ning ;
Yuan, Xiaojing ;
Chen, Ji ;
Zouridakis, George .
2008 30TH ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY, VOLS 1-8, 2008, :3110-+
[46]   Video Google: A text retrieval approach to object matching in videos [J].
Sivic, J ;
Zisserman, A .
NINTH IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION, VOLS I AND II, PROCEEDINGS, 2003, :1470-+
[47]   A relative color approach to color discrimination for malignant melanoma detection in dermoscopy images [J].
Stanley, R. Joe ;
Stoecker, William V. ;
Moss, Randy H. .
SKIN RESEARCH AND TECHNOLOGY, 2007, 13 (01) :62-72
[48]   Detection of granularity in dermoscopy images of malignant melanoma using color and texture features [J].
Stoecker, William V. ;
Wronkiewiecz, Mark ;
Chowdhury, Raeed ;
Stanley, R. Joe ;
Xu, Jin ;
Bangert, Austin ;
Shrestha, Bijaya ;
Calcara, David A. ;
Rabinovitz, Harold S. ;
Oliviero, Margaret ;
Ahmed, Fatimah ;
Perry, Lindall A. ;
Drugge, Rhett .
COMPUTERIZED MEDICAL IMAGING AND GRAPHICS, 2011, 35 (02) :144-147
[49]   Detection of asymmetric blotches (asymmetric structureless areas) in dermoscopy images of malignant melanoma using relative color [J].
Stoecker, WV ;
Gupta, K ;
Stanley, RJ ;
Moss, RH ;
Shrestha, B .
SKIN RESEARCH AND TECHNOLOGY, 2005, 11 (03) :179-184
[50]  
STOLZ W, 1994, EUR J DERMATOL, V4, P521