Non-melanoma skin lesion classification using colour image data in a hierarchical K-NN classifier

Lucia Ballerini; Robert B. Fisher; Ben Aldridge; Jonathan Rees

doi:10.1109/ISBI.2012.6235558

Non-melanoma skin lesion classification using colour image data in a hierarchical K-NN classifier

Lucia Ballerini, Robert B. Fisher, Ben Aldridge, Jonathan Rees

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

28 Citations (Scopus)

Abstract

This paper presents an algorithm for classification of non-melanoma skin lesions based on a novel hierarchical K-Nearest Neighbors (K-NN) classifier. The K-NN classifier is simple, quick and effective. The hierarchical structure decomposes the classification task into a set of simpler problems, one at each node of the classification. Feature selection is embedded in the hierarchical framework that chooses the most relevant feature subsets at each node of the hierarchy. Colour and texture features are extracted from skin lesions. The accuracy of the proposed hierarchical scheme is higher than 93% in discriminating cancer and pre-malignant lesions from benign lesions, and it reaches an overall classification accuracy of 74% over five common classes of skin lesions, including two non-melanoma cancer types. This is the most extensive published result on non-melanoma skin cancer classification from colour images acquired by a standard camera (non-dermoscopy).

Original language	English
Title of host publication	2012 9th IEEE International Symposium on Biomedical Imaging
Subtitle of host publication	From Nano to Macro - Proceedings
Place of Publication	Piscataway
Publisher	IEEE
Pages	358-361
Number of pages	4
ISBN (Print)	9781457718588
DOIs	https://doi.org/10.1109/ISBI.2012.6235558
Publication status	Published - 2012
Event	9th IEEE International Symposium on Biomedical Imaging: From Nano to Macro - Centre de Convencions Internacional de Barcelona, Barcelona, Spain Duration: 2 May 2012 → 5 May 2012

Conference

Conference	9th IEEE International Symposium on Biomedical Imaging: From Nano to Macro
Abbreviated title	ISBI 2012
Country/Territory	Spain
City	Barcelona
Period	2/05/12 → 5/05/12

UN SDGs

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

Access to Document

10.1109/ISBI.2012.6235558

Cite this

@inproceedings{cc949c7c989f4a97916bdaee21413061,

title = "Non-melanoma skin lesion classification using colour image data in a hierarchical K-NN classifier",

abstract = "This paper presents an algorithm for classification of non-melanoma skin lesions based on a novel hierarchical K-Nearest Neighbors (K-NN) classifier. The K-NN classifier is simple, quick and effective. The hierarchical structure decomposes the classification task into a set of simpler problems, one at each node of the classification. Feature selection is embedded in the hierarchical framework that chooses the most relevant feature subsets at each node of the hierarchy. Colour and texture features are extracted from skin lesions. The accuracy of the proposed hierarchical scheme is higher than 93% in discriminating cancer and pre-malignant lesions from benign lesions, and it reaches an overall classification accuracy of 74% over five common classes of skin lesions, including two non-melanoma cancer types. This is the most extensive published result on non-melanoma skin cancer classification from colour images acquired by a standard camera (non-dermoscopy).",

author = "Lucia Ballerini and Fisher, {Robert B.} and Ben Aldridge and Jonathan Rees",

year = "2012",

doi = "10.1109/ISBI.2012.6235558",

language = "English",

isbn = "9781457718588",

pages = "358--361",

booktitle = "2012 9th IEEE International Symposium on Biomedical Imaging",

publisher = "IEEE",

note = "9th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2012 ; Conference date: 02-05-2012 Through 05-05-2012",

}

Ballerini, L, Fisher, RB, Aldridge, B & Rees, J 2012, Non-melanoma skin lesion classification using colour image data in a hierarchical K-NN classifier. in 2012 9th IEEE International Symposium on Biomedical Imaging: From Nano to Macro - Proceedings. IEEE, Piscataway, pp. 358-361, 9th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, Barcelona, Spain, 2/05/12. https://doi.org/10.1109/ISBI.2012.6235558

Non-melanoma skin lesion classification using colour image data in a hierarchical K-NN classifier. / Ballerini, Lucia; Fisher, Robert B.; Aldridge, Ben et al.
2012 9th IEEE International Symposium on Biomedical Imaging: From Nano to Macro - Proceedings. Piscataway: IEEE, 2012. p. 358-361.

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

TY - GEN

T1 - Non-melanoma skin lesion classification using colour image data in a hierarchical K-NN classifier

AU - Ballerini, Lucia

AU - Fisher, Robert B.

AU - Aldridge, Ben

AU - Rees, Jonathan

PY - 2012

Y1 - 2012

N2 - This paper presents an algorithm for classification of non-melanoma skin lesions based on a novel hierarchical K-Nearest Neighbors (K-NN) classifier. The K-NN classifier is simple, quick and effective. The hierarchical structure decomposes the classification task into a set of simpler problems, one at each node of the classification. Feature selection is embedded in the hierarchical framework that chooses the most relevant feature subsets at each node of the hierarchy. Colour and texture features are extracted from skin lesions. The accuracy of the proposed hierarchical scheme is higher than 93% in discriminating cancer and pre-malignant lesions from benign lesions, and it reaches an overall classification accuracy of 74% over five common classes of skin lesions, including two non-melanoma cancer types. This is the most extensive published result on non-melanoma skin cancer classification from colour images acquired by a standard camera (non-dermoscopy).

AB - This paper presents an algorithm for classification of non-melanoma skin lesions based on a novel hierarchical K-Nearest Neighbors (K-NN) classifier. The K-NN classifier is simple, quick and effective. The hierarchical structure decomposes the classification task into a set of simpler problems, one at each node of the classification. Feature selection is embedded in the hierarchical framework that chooses the most relevant feature subsets at each node of the hierarchy. Colour and texture features are extracted from skin lesions. The accuracy of the proposed hierarchical scheme is higher than 93% in discriminating cancer and pre-malignant lesions from benign lesions, and it reaches an overall classification accuracy of 74% over five common classes of skin lesions, including two non-melanoma cancer types. This is the most extensive published result on non-melanoma skin cancer classification from colour images acquired by a standard camera (non-dermoscopy).

UR - http://www.scopus.com/inward/record.url?scp=84864863492&partnerID=8YFLogxK

U2 - 10.1109/ISBI.2012.6235558

DO - 10.1109/ISBI.2012.6235558

M3 - Conference contribution

AN - SCOPUS:84864863492

SN - 9781457718588

SP - 358

EP - 361

BT - 2012 9th IEEE International Symposium on Biomedical Imaging

PB - IEEE

CY - Piscataway

T2 - 9th IEEE International Symposium on Biomedical Imaging: From Nano to Macro

Y2 - 2 May 2012 through 5 May 2012

ER -

Non-melanoma skin lesion classification using colour image data in a hierarchical K-NN classifier

Abstract

Conference

UN SDGs

Access to Document

Other files and links

Fingerprint