Scholars' Mine
Missouri S&T
Research Repository
Curtis Laws Wilson Library
400 W. 14th Street
Rolla, MO 65409-0060
scholarsmine@mst.edu
| Title: | A methodological approach to the classification of dermoscopy images |
| Author (s): | Celebi, M. Emre Kingravi, Hassan A. Uddin, Bakhtiyar Iyatomi, Hitoshi Aslandogan, Y. Alp Stoecker, William V. Moss, Randy Hays |
| Department/Lab Affiliations: | Electrical and Computer Engineering Image Processing Laboratory |
| Keywords: | dermoscopy model selection support vector machine |
| Subject Terms: | Machine learning. Skin -- Cancer. |
| Issue Date: | 2007 |
| Publisher: | Elsevier |
| Citation: | Celebi, M.Emre, Hassan A. Kingravi, Bakhtiyar Uddin, Hitoshi Iyatomi, Y. Alp Aslandogan, William V. Stoecker, Randy H. Moss, "A Methodological Approach to the Classification of Dermoscopy Images", Computerized Medical Imaging and Graphics, Vol. 31, no. 6, pp. 362-373, 2007. |
| Abstract: | In this paper a methodological approach to the classification of pigmented skin lesions in dermoscopy images is presented. First, automatic border detection is performed to separate the lesion from the background skin. Shape features are then extracted from this border. For the extraction of color and texture related features, the image is divided into various clinically significant regions using the Euclidean distance transform. This feature data is fed into an optimization framework, which ranks the features using various feature selection algorithms and determines the optimal feature subset size according to the area under the ROC curve measure obtained from support vector machine classification. The issue of class imbalance is addressed using various sampling strategies, and the classifier generalization error is estimated using Monte Carlo cross validation. Experiments on a set of 564 images yielded a specificity of 92.34% and a sensitivity of 93.33%. |
| Type: | Article - Journal text |
| In Title: | Computerized Medical Imaging and Graphics |
| Copyright Notice: | Pre-print: author can archive with restrictions;Restriction: This does not include Cell Press; Post-print: author can archive; This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder. FULL COPYRIGHT INFORMATION: |
| Publisher URL: | |
| Link to this page: |
| title | A methodological approach to the classification of dermoscopy images |
| contributor.author | Celebi, M. Emre |
| contributor.author | Kingravi, Hassan A. |
| contributor.author | Uddin, Bakhtiyar |
| contributor.author | Iyatomi, Hitoshi |
| contributor.author | Aslandogan, Y. Alp |
| contributor.author | Stoecker, William V. |
| contributor.author | Moss, Randy Hays |
| contributor.deptlab | Electrical and Computer Engineering |
| contributor.deptlab | Image Processing Laboratory |
| contributor.sponsor | EDRA Interactive Atlas of Dermoscopy |
| contributor.sponsor | James A. Schlipmann Melanoma Cancer Foundation |
| contributor.sponsor | National Institute of Health |
| contributor.sponsor | National Science Foundation |
| contributor.sponsor | Texas Workforce Commission |
| subject | dermoscopy |
| subject | model selection |
| subject | support vector machine |
| subject.LCC | Melanoma. |
| subject.LCSH | Machine learning. |
| subject.LCSH | Skin -- Cancer. |
| date.issued | 2007 |
| publisher | Elsevier |
| identifier.citation | Celebi, M.Emre, Hassan A. Kingravi, Bakhtiyar Uddin, Hitoshi Iyatomi, Y. Alp Aslandogan, William V. Stoecker, Randy H. Moss, "A Methodological Approach to the Classification of Dermoscopy Images", Computerized Medical Imaging and Graphics, Vol. 31, no. 6, pp. 362-373, 2007. |
| identifier.pub.URI | |
| description.abstract | In this paper a methodological approach to the classification of pigmented skin lesions in dermoscopy images is presented. First, automatic border detection is performed to separate the lesion from the background skin. Shape features are then extracted from this border. For the extraction of color and texture related features, the image is divided into various clinically significant regions using the Euclidean distance transform. This feature data is fed into an optimization framework, which ranks the features using various feature selection algorithms and determines the optimal feature subset size according to the area under the ROC curve measure obtained from support vector machine classification. The issue of class imbalance is addressed using various sampling strategies, and the classifier generalization error is estimated using Monte Carlo cross validation. Experiments on a set of 564 images yielded a specificity of 92.34% and a sensitivity of 93.33%. |
| type | Article - Journal |
| type.DCMIType | text |
| type.status | Postprint |
| rights | Pre-print: author can archive with restrictions;Restriction: This does not include Cell Press; Post-print: author can archive; |
| rights | This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder. |
| rights.URI | |
| relation.isPartOf | Computerized Medical Imaging and Graphics |
| date.available | 2008-07-07T19:47:05Z |
| identifier.persist.URI |