Mining biomedical images with density-based clustering

M. Emre Celebi, Y. Alp Aslandogan, Paul R. Bergstresser

Research output: Chapter in Book/Report/Conference proceedingConference contribution

30 Citations (Scopus)

Abstract

Density-based clustering algorithms have recently gained popularity in the data mining field due to their ability to discover arbitrary shaped clusters while preserving spatial proximity of data points. In this work we adapt a density-based clustering algorithm, DBSCAN, to a new problem domain: Identification of homogenous color regions in biomedical images. Examples of specific problems of this nature include landscape segmentation of satellite imagery, object detection and, in our case, identification of significant color regions in images of skin lesions (tumors). Automated outer and inner boundary segmentation is a key step in segmentation of structures such as skin lesions, tumors of breast, bone, and brain. This step is important because the accuracy of the subsequent steps (extraction of various features, post-processing) crucially depends on the accuracy of this very first step. In this paper, we present an unsupervised approach to segmentation of pigmented skin lesion images based on DBSCAN clustering algorithm. The color regions identified by the algorithm are compared to those identified by the human subjects and the Kappa coefficient, a statistical indicator of computer-human agreement, is found to be significant.

Original languageEnglish (US)
Title of host publicationInternational Conference on Information Technology: Coding and Computing, ITCC
EditorsH. Selvaraj, P.K. Srimani
Pages163-168
Number of pages6
Volume1
StatePublished - 2005
EventITCC 2005 - International Conference on Information Technology: Coding and Computing - Las Vegas, NV, United States
Duration: Apr 4 2005Apr 6 2005

Other

OtherITCC 2005 - International Conference on Information Technology: Coding and Computing
CountryUnited States
CityLas Vegas, NV
Period4/4/054/6/05

Fingerprint

Clustering algorithms
Skin
Color
Tumors
Satellite imagery
Data mining
Brain
Bone
Processing

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Celebi, M. E., Aslandogan, Y. A., & Bergstresser, P. R. (2005). Mining biomedical images with density-based clustering. In H. Selvaraj, & P. K. Srimani (Eds.), International Conference on Information Technology: Coding and Computing, ITCC (Vol. 1, pp. 163-168)

Mining biomedical images with density-based clustering. / Celebi, M. Emre; Aslandogan, Y. Alp; Bergstresser, Paul R.

International Conference on Information Technology: Coding and Computing, ITCC. ed. / H. Selvaraj; P.K. Srimani. Vol. 1 2005. p. 163-168.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Celebi, ME, Aslandogan, YA & Bergstresser, PR 2005, Mining biomedical images with density-based clustering. in H Selvaraj & PK Srimani (eds), International Conference on Information Technology: Coding and Computing, ITCC. vol. 1, pp. 163-168, ITCC 2005 - International Conference on Information Technology: Coding and Computing, Las Vegas, NV, United States, 4/4/05.
Celebi ME, Aslandogan YA, Bergstresser PR. Mining biomedical images with density-based clustering. In Selvaraj H, Srimani PK, editors, International Conference on Information Technology: Coding and Computing, ITCC. Vol. 1. 2005. p. 163-168
Celebi, M. Emre ; Aslandogan, Y. Alp ; Bergstresser, Paul R. / Mining biomedical images with density-based clustering. International Conference on Information Technology: Coding and Computing, ITCC. editor / H. Selvaraj ; P.K. Srimani. Vol. 1 2005. pp. 163-168
@inproceedings{6ba84634b4844939aca0e7f9e2d0c433,
title = "Mining biomedical images with density-based clustering",
abstract = "Density-based clustering algorithms have recently gained popularity in the data mining field due to their ability to discover arbitrary shaped clusters while preserving spatial proximity of data points. In this work we adapt a density-based clustering algorithm, DBSCAN, to a new problem domain: Identification of homogenous color regions in biomedical images. Examples of specific problems of this nature include landscape segmentation of satellite imagery, object detection and, in our case, identification of significant color regions in images of skin lesions (tumors). Automated outer and inner boundary segmentation is a key step in segmentation of structures such as skin lesions, tumors of breast, bone, and brain. This step is important because the accuracy of the subsequent steps (extraction of various features, post-processing) crucially depends on the accuracy of this very first step. In this paper, we present an unsupervised approach to segmentation of pigmented skin lesion images based on DBSCAN clustering algorithm. The color regions identified by the algorithm are compared to those identified by the human subjects and the Kappa coefficient, a statistical indicator of computer-human agreement, is found to be significant.",
author = "Celebi, {M. Emre} and Aslandogan, {Y. Alp} and Bergstresser, {Paul R.}",
year = "2005",
language = "English (US)",
isbn = "0769523153",
volume = "1",
pages = "163--168",
editor = "H. Selvaraj and P.K. Srimani",
booktitle = "International Conference on Information Technology: Coding and Computing, ITCC",

}

TY - GEN

T1 - Mining biomedical images with density-based clustering

AU - Celebi, M. Emre

AU - Aslandogan, Y. Alp

AU - Bergstresser, Paul R.

PY - 2005

Y1 - 2005

N2 - Density-based clustering algorithms have recently gained popularity in the data mining field due to their ability to discover arbitrary shaped clusters while preserving spatial proximity of data points. In this work we adapt a density-based clustering algorithm, DBSCAN, to a new problem domain: Identification of homogenous color regions in biomedical images. Examples of specific problems of this nature include landscape segmentation of satellite imagery, object detection and, in our case, identification of significant color regions in images of skin lesions (tumors). Automated outer and inner boundary segmentation is a key step in segmentation of structures such as skin lesions, tumors of breast, bone, and brain. This step is important because the accuracy of the subsequent steps (extraction of various features, post-processing) crucially depends on the accuracy of this very first step. In this paper, we present an unsupervised approach to segmentation of pigmented skin lesion images based on DBSCAN clustering algorithm. The color regions identified by the algorithm are compared to those identified by the human subjects and the Kappa coefficient, a statistical indicator of computer-human agreement, is found to be significant.

AB - Density-based clustering algorithms have recently gained popularity in the data mining field due to their ability to discover arbitrary shaped clusters while preserving spatial proximity of data points. In this work we adapt a density-based clustering algorithm, DBSCAN, to a new problem domain: Identification of homogenous color regions in biomedical images. Examples of specific problems of this nature include landscape segmentation of satellite imagery, object detection and, in our case, identification of significant color regions in images of skin lesions (tumors). Automated outer and inner boundary segmentation is a key step in segmentation of structures such as skin lesions, tumors of breast, bone, and brain. This step is important because the accuracy of the subsequent steps (extraction of various features, post-processing) crucially depends on the accuracy of this very first step. In this paper, we present an unsupervised approach to segmentation of pigmented skin lesion images based on DBSCAN clustering algorithm. The color regions identified by the algorithm are compared to those identified by the human subjects and the Kappa coefficient, a statistical indicator of computer-human agreement, is found to be significant.

UR - http://www.scopus.com/inward/record.url?scp=24744450986&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=24744450986&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:24744450986

SN - 0769523153

VL - 1

SP - 163

EP - 168

BT - International Conference on Information Technology: Coding and Computing, ITCC

A2 - Selvaraj, H.

A2 - Srimani, P.K.

ER -