Spectral Clustering for Unsupervised Segmentation of Lower Extremity Wound Beds Using Optical Images

Dhiraj Manohar Dhane, Vishal Krishna, Arun Achar, Chittaranjan Bar, Kunal Sanyal, Chandan Chakraborty

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

Chronic lower extremity wound is a complicated disease condition of localized injury to skin and its tissues which have plagued many elders worldwide. The ulcer assessment and management is expensive and is burden on health establishment. Currently accurate wound evaluation remains a tedious task as it rely on visual inspection. This paper propose a new method for wound-area detection, using images digitally captured by a hand-held, optical camera. The strategy proposed involves spectral approach for clustering, based on the affinity matrix. The spectral clustering (SC) involves construction of similarity matrix of Laplacian based on Ng-Jorden-Weiss algorithm. Starting with a quadratic method, wound photographs were pre-processed for color homogenization. The first-order statistics filter was then applied to extract spurious regions. The filter was selected based on the performance, evaluated on four quality metrics. Then, the spectral method was used on the filtered images for effective segmentation. The segmented regions were post-processed using morphological operators. The performance of spectral segmentation was confirmed by ground-truth pictures labeled by dermatologists. The SC results were additionally compared with the results of k-means and Fuzzy C-Means (FCM) clustering algorithms. The SC approach on a set of 105 images, effectively delineated targeted wound beds yielding a segmentation accuracy of 86.73 %, positive predictive values of 91.80 %, and a sensitivity of 89.54 %. This approach shows the robustness of tool for ulcer perimeter measurement and healing progression. The article elucidates its potential to be incorporated in patient facing medical systems targeting a rapid clinical assistance.

Original languageEnglish
Article number207
JournalJournal of Medical Systems
Volume40
Issue number9
DOIs
Publication statusPublished - 01-09-2016
Externally publishedYes

Fingerprint

Cluster Analysis
Lower Extremity
Wounds and Injuries
Clustering algorithms
Skin
Inspection
Cameras
Health
Statistics
Tissue
Color
Ulcer
Hand

All Science Journal Classification (ASJC) codes

  • Medicine (miscellaneous)
  • Information Systems
  • Health Informatics
  • Health Information Management

Cite this

Dhane, Dhiraj Manohar ; Krishna, Vishal ; Achar, Arun ; Bar, Chittaranjan ; Sanyal, Kunal ; Chakraborty, Chandan. / Spectral Clustering for Unsupervised Segmentation of Lower Extremity Wound Beds Using Optical Images. In: Journal of Medical Systems. 2016 ; Vol. 40, No. 9.
@article{3959b31ce4564eb580c96ad23491a2a6,
title = "Spectral Clustering for Unsupervised Segmentation of Lower Extremity Wound Beds Using Optical Images",
abstract = "Chronic lower extremity wound is a complicated disease condition of localized injury to skin and its tissues which have plagued many elders worldwide. The ulcer assessment and management is expensive and is burden on health establishment. Currently accurate wound evaluation remains a tedious task as it rely on visual inspection. This paper propose a new method for wound-area detection, using images digitally captured by a hand-held, optical camera. The strategy proposed involves spectral approach for clustering, based on the affinity matrix. The spectral clustering (SC) involves construction of similarity matrix of Laplacian based on Ng-Jorden-Weiss algorithm. Starting with a quadratic method, wound photographs were pre-processed for color homogenization. The first-order statistics filter was then applied to extract spurious regions. The filter was selected based on the performance, evaluated on four quality metrics. Then, the spectral method was used on the filtered images for effective segmentation. The segmented regions were post-processed using morphological operators. The performance of spectral segmentation was confirmed by ground-truth pictures labeled by dermatologists. The SC results were additionally compared with the results of k-means and Fuzzy C-Means (FCM) clustering algorithms. The SC approach on a set of 105 images, effectively delineated targeted wound beds yielding a segmentation accuracy of 86.73 {\%}, positive predictive values of 91.80 {\%}, and a sensitivity of 89.54 {\%}. This approach shows the robustness of tool for ulcer perimeter measurement and healing progression. The article elucidates its potential to be incorporated in patient facing medical systems targeting a rapid clinical assistance.",
author = "Dhane, {Dhiraj Manohar} and Vishal Krishna and Arun Achar and Chittaranjan Bar and Kunal Sanyal and Chandan Chakraborty",
year = "2016",
month = "9",
day = "1",
doi = "10.1007/s10916-016-0554-x",
language = "English",
volume = "40",
journal = "Journal of Medical Systems",
issn = "0148-5598",
publisher = "Springer New York",
number = "9",

}

Spectral Clustering for Unsupervised Segmentation of Lower Extremity Wound Beds Using Optical Images. / Dhane, Dhiraj Manohar; Krishna, Vishal; Achar, Arun; Bar, Chittaranjan; Sanyal, Kunal; Chakraborty, Chandan.

In: Journal of Medical Systems, Vol. 40, No. 9, 207, 01.09.2016.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Spectral Clustering for Unsupervised Segmentation of Lower Extremity Wound Beds Using Optical Images

AU - Dhane, Dhiraj Manohar

AU - Krishna, Vishal

AU - Achar, Arun

AU - Bar, Chittaranjan

AU - Sanyal, Kunal

AU - Chakraborty, Chandan

PY - 2016/9/1

Y1 - 2016/9/1

N2 - Chronic lower extremity wound is a complicated disease condition of localized injury to skin and its tissues which have plagued many elders worldwide. The ulcer assessment and management is expensive and is burden on health establishment. Currently accurate wound evaluation remains a tedious task as it rely on visual inspection. This paper propose a new method for wound-area detection, using images digitally captured by a hand-held, optical camera. The strategy proposed involves spectral approach for clustering, based on the affinity matrix. The spectral clustering (SC) involves construction of similarity matrix of Laplacian based on Ng-Jorden-Weiss algorithm. Starting with a quadratic method, wound photographs were pre-processed for color homogenization. The first-order statistics filter was then applied to extract spurious regions. The filter was selected based on the performance, evaluated on four quality metrics. Then, the spectral method was used on the filtered images for effective segmentation. The segmented regions were post-processed using morphological operators. The performance of spectral segmentation was confirmed by ground-truth pictures labeled by dermatologists. The SC results were additionally compared with the results of k-means and Fuzzy C-Means (FCM) clustering algorithms. The SC approach on a set of 105 images, effectively delineated targeted wound beds yielding a segmentation accuracy of 86.73 %, positive predictive values of 91.80 %, and a sensitivity of 89.54 %. This approach shows the robustness of tool for ulcer perimeter measurement and healing progression. The article elucidates its potential to be incorporated in patient facing medical systems targeting a rapid clinical assistance.

AB - Chronic lower extremity wound is a complicated disease condition of localized injury to skin and its tissues which have plagued many elders worldwide. The ulcer assessment and management is expensive and is burden on health establishment. Currently accurate wound evaluation remains a tedious task as it rely on visual inspection. This paper propose a new method for wound-area detection, using images digitally captured by a hand-held, optical camera. The strategy proposed involves spectral approach for clustering, based on the affinity matrix. The spectral clustering (SC) involves construction of similarity matrix of Laplacian based on Ng-Jorden-Weiss algorithm. Starting with a quadratic method, wound photographs were pre-processed for color homogenization. The first-order statistics filter was then applied to extract spurious regions. The filter was selected based on the performance, evaluated on four quality metrics. Then, the spectral method was used on the filtered images for effective segmentation. The segmented regions were post-processed using morphological operators. The performance of spectral segmentation was confirmed by ground-truth pictures labeled by dermatologists. The SC results were additionally compared with the results of k-means and Fuzzy C-Means (FCM) clustering algorithms. The SC approach on a set of 105 images, effectively delineated targeted wound beds yielding a segmentation accuracy of 86.73 %, positive predictive values of 91.80 %, and a sensitivity of 89.54 %. This approach shows the robustness of tool for ulcer perimeter measurement and healing progression. The article elucidates its potential to be incorporated in patient facing medical systems targeting a rapid clinical assistance.

UR - http://www.scopus.com/inward/record.url?scp=84982261587&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84982261587&partnerID=8YFLogxK

U2 - 10.1007/s10916-016-0554-x

DO - 10.1007/s10916-016-0554-x

M3 - Article

VL - 40

JO - Journal of Medical Systems

JF - Journal of Medical Systems

SN - 0148-5598

IS - 9

M1 - 207

ER -