A Minimum Spanning Forest-Based Method for Noninvasive Cancer Detection with Hyperspectral Imaging

Robert Pike, Guolan Lu, Dongsheng Wang, Zhuo Georgia Chen, Baowei Fei

Research output: Contribution to journalArticle

24 Citations (Scopus)

Abstract

Goal: The purpose of this paper is to develop a classification method that combines both spectral and spatial information for distinguishing cancer from healthy tissue on hyperspectral images in an animal model. Methods: An automated algorithm based on a minimum spanning forest (MSF) and optimal band selection has been proposed to classify healthy and cancerous tissue on hyperspectral images. A support vector machine classifier is trained to create a pixel-wise classification probability map of cancerous and healthy tissue. This map is then used to identify markers that are used to compute mutual information for a range of bands in the hyperspectral image and thus select the optimal bands. An MSF is finally grown to segment the image using spatial and spectral information. Conclusion: The MSF based method with automatically selected bands proved to be accurate in determining the tumor boundary on hyperspectral images. Significance: Hyperspectral imaging combined with the proposed classification technique has the potential to provide a noninvasive tool for cancer detection.

Original languageEnglish (US)
Article number7202847
Pages (from-to)653-663
Number of pages11
JournalIEEE Transactions on Biomedical Engineering
Volume63
Issue number3
DOIs
StatePublished - Mar 2016
Externally publishedYes

Fingerprint

Tissue
Support vector machines
Tumors
Animals
Classifiers
Pixels
Hyperspectral imaging

Keywords

  • Hyperspectral imaging
  • image classification
  • minimum spanning forest
  • mutual information
  • noninvasive cancer detection
  • support vector machine

ASJC Scopus subject areas

  • Biomedical Engineering

Cite this

A Minimum Spanning Forest-Based Method for Noninvasive Cancer Detection with Hyperspectral Imaging. / Pike, Robert; Lu, Guolan; Wang, Dongsheng; Chen, Zhuo Georgia; Fei, Baowei.

In: IEEE Transactions on Biomedical Engineering, Vol. 63, No. 3, 7202847, 03.2016, p. 653-663.

Research output: Contribution to journalArticle

Pike, Robert ; Lu, Guolan ; Wang, Dongsheng ; Chen, Zhuo Georgia ; Fei, Baowei. / A Minimum Spanning Forest-Based Method for Noninvasive Cancer Detection with Hyperspectral Imaging. In: IEEE Transactions on Biomedical Engineering. 2016 ; Vol. 63, No. 3. pp. 653-663.
@article{8d7af5916a50408e888b1184150da914,
title = "A Minimum Spanning Forest-Based Method for Noninvasive Cancer Detection with Hyperspectral Imaging",
abstract = "Goal: The purpose of this paper is to develop a classification method that combines both spectral and spatial information for distinguishing cancer from healthy tissue on hyperspectral images in an animal model. Methods: An automated algorithm based on a minimum spanning forest (MSF) and optimal band selection has been proposed to classify healthy and cancerous tissue on hyperspectral images. A support vector machine classifier is trained to create a pixel-wise classification probability map of cancerous and healthy tissue. This map is then used to identify markers that are used to compute mutual information for a range of bands in the hyperspectral image and thus select the optimal bands. An MSF is finally grown to segment the image using spatial and spectral information. Conclusion: The MSF based method with automatically selected bands proved to be accurate in determining the tumor boundary on hyperspectral images. Significance: Hyperspectral imaging combined with the proposed classification technique has the potential to provide a noninvasive tool for cancer detection.",
keywords = "Hyperspectral imaging, image classification, minimum spanning forest, mutual information, noninvasive cancer detection, support vector machine",
author = "Robert Pike and Guolan Lu and Dongsheng Wang and Chen, {Zhuo Georgia} and Baowei Fei",
year = "2016",
month = "3",
doi = "10.1109/TBME.2015.2468578",
language = "English (US)",
volume = "63",
pages = "653--663",
journal = "IEEE Transactions on Biomedical Engineering",
issn = "0018-9294",
publisher = "IEEE Computer Society",
number = "3",

}

TY - JOUR

T1 - A Minimum Spanning Forest-Based Method for Noninvasive Cancer Detection with Hyperspectral Imaging

AU - Pike, Robert

AU - Lu, Guolan

AU - Wang, Dongsheng

AU - Chen, Zhuo Georgia

AU - Fei, Baowei

PY - 2016/3

Y1 - 2016/3

N2 - Goal: The purpose of this paper is to develop a classification method that combines both spectral and spatial information for distinguishing cancer from healthy tissue on hyperspectral images in an animal model. Methods: An automated algorithm based on a minimum spanning forest (MSF) and optimal band selection has been proposed to classify healthy and cancerous tissue on hyperspectral images. A support vector machine classifier is trained to create a pixel-wise classification probability map of cancerous and healthy tissue. This map is then used to identify markers that are used to compute mutual information for a range of bands in the hyperspectral image and thus select the optimal bands. An MSF is finally grown to segment the image using spatial and spectral information. Conclusion: The MSF based method with automatically selected bands proved to be accurate in determining the tumor boundary on hyperspectral images. Significance: Hyperspectral imaging combined with the proposed classification technique has the potential to provide a noninvasive tool for cancer detection.

AB - Goal: The purpose of this paper is to develop a classification method that combines both spectral and spatial information for distinguishing cancer from healthy tissue on hyperspectral images in an animal model. Methods: An automated algorithm based on a minimum spanning forest (MSF) and optimal band selection has been proposed to classify healthy and cancerous tissue on hyperspectral images. A support vector machine classifier is trained to create a pixel-wise classification probability map of cancerous and healthy tissue. This map is then used to identify markers that are used to compute mutual information for a range of bands in the hyperspectral image and thus select the optimal bands. An MSF is finally grown to segment the image using spatial and spectral information. Conclusion: The MSF based method with automatically selected bands proved to be accurate in determining the tumor boundary on hyperspectral images. Significance: Hyperspectral imaging combined with the proposed classification technique has the potential to provide a noninvasive tool for cancer detection.

KW - Hyperspectral imaging

KW - image classification

KW - minimum spanning forest

KW - mutual information

KW - noninvasive cancer detection

KW - support vector machine

UR - http://www.scopus.com/inward/record.url?scp=84962090392&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84962090392&partnerID=8YFLogxK

U2 - 10.1109/TBME.2015.2468578

DO - 10.1109/TBME.2015.2468578

M3 - Article

C2 - 26285052

AN - SCOPUS:84962090392

VL - 63

SP - 653

EP - 663

JO - IEEE Transactions on Biomedical Engineering

JF - IEEE Transactions on Biomedical Engineering

SN - 0018-9294

IS - 3

M1 - 7202847

ER -