Hair detection and lesion segmentation in dermoscopic images using domain knowledge

Sameena Pathan, K. Gopalakrishna Prabhu, P. C. Siddalingaswamy

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Automated segmentation and dermoscopic hair detection are one of the significant challenges in computer-aided diagnosis (CAD) of melanocytic lesions. Additionally, due to the presence of artifacts and variation in skin texture and smooth lesion boundaries, the accuracy of such methods gets hampered. The objective of this research is to develop an automated hair detection and lesion segmentation algorithm using lesion-specific properties to improve the accuracy. The aforementioned objective is achieved in two ways. Firstly, a novel hair detection algorithm is designed by considering the properties of dermoscopic hair. Second, a novel chroma-based geometric deformable model is used to effectively differentiate the lesion from the surrounding skin. The speed function incorporates the chrominance properties of the lesion to stop evolution at the lesion boundary. Automatic initialization of the initial contour and chrominance-based speed function aids in providing robust and flexible segmentation. The proposed approach is tested on 200 images from PH2 and 900 images from ISBI 2016 datasets. Average accuracy, sensitivity, specificity, and overlap scores of 93.4, 87.6, 95.3, and 11.52% respectively are obtained for the PH2 dataset. Similarly, the proposed method resulted in average accuracy, sensitivity, specificity, and overlap scores of 94.6, 82.4, 97.2, and 7.20% respectively for the ISBI 2016 dataset. Statistical and quantitative analyses prove the reliability of the algorithm for incorporation in CAD systems. [Figure not available: see fulltext.].

Original languageEnglish
Pages (from-to)2051-2065
Number of pages15
JournalMedical and Biological Engineering and Computing
Volume56
Issue number11
DOIs
Publication statusPublished - 01-11-2018

Fingerprint

Computer aided diagnosis
Skin
Textures

All Science Journal Classification (ASJC) codes

  • Biomedical Engineering
  • Computer Science Applications

Cite this

@article{040b78cd68284470a3aa7464a65d298e,
title = "Hair detection and lesion segmentation in dermoscopic images using domain knowledge",
abstract = "Automated segmentation and dermoscopic hair detection are one of the significant challenges in computer-aided diagnosis (CAD) of melanocytic lesions. Additionally, due to the presence of artifacts and variation in skin texture and smooth lesion boundaries, the accuracy of such methods gets hampered. The objective of this research is to develop an automated hair detection and lesion segmentation algorithm using lesion-specific properties to improve the accuracy. The aforementioned objective is achieved in two ways. Firstly, a novel hair detection algorithm is designed by considering the properties of dermoscopic hair. Second, a novel chroma-based geometric deformable model is used to effectively differentiate the lesion from the surrounding skin. The speed function incorporates the chrominance properties of the lesion to stop evolution at the lesion boundary. Automatic initialization of the initial contour and chrominance-based speed function aids in providing robust and flexible segmentation. The proposed approach is tested on 200 images from PH2 and 900 images from ISBI 2016 datasets. Average accuracy, sensitivity, specificity, and overlap scores of 93.4, 87.6, 95.3, and 11.52{\%} respectively are obtained for the PH2 dataset. Similarly, the proposed method resulted in average accuracy, sensitivity, specificity, and overlap scores of 94.6, 82.4, 97.2, and 7.20{\%} respectively for the ISBI 2016 dataset. Statistical and quantitative analyses prove the reliability of the algorithm for incorporation in CAD systems. [Figure not available: see fulltext.].",
author = "Sameena Pathan and Prabhu, {K. Gopalakrishna} and Siddalingaswamy, {P. C.}",
year = "2018",
month = "11",
day = "1",
doi = "10.1007/s11517-018-1837-9",
language = "English",
volume = "56",
pages = "2051--2065",
journal = "Medical and Biological Engineering and Computing",
issn = "0140-0118",
publisher = "Springer Verlag",
number = "11",

}

Hair detection and lesion segmentation in dermoscopic images using domain knowledge. / Pathan, Sameena; Prabhu, K. Gopalakrishna; Siddalingaswamy, P. C.

In: Medical and Biological Engineering and Computing, Vol. 56, No. 11, 01.11.2018, p. 2051-2065.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Hair detection and lesion segmentation in dermoscopic images using domain knowledge

AU - Pathan, Sameena

AU - Prabhu, K. Gopalakrishna

AU - Siddalingaswamy, P. C.

PY - 2018/11/1

Y1 - 2018/11/1

N2 - Automated segmentation and dermoscopic hair detection are one of the significant challenges in computer-aided diagnosis (CAD) of melanocytic lesions. Additionally, due to the presence of artifacts and variation in skin texture and smooth lesion boundaries, the accuracy of such methods gets hampered. The objective of this research is to develop an automated hair detection and lesion segmentation algorithm using lesion-specific properties to improve the accuracy. The aforementioned objective is achieved in two ways. Firstly, a novel hair detection algorithm is designed by considering the properties of dermoscopic hair. Second, a novel chroma-based geometric deformable model is used to effectively differentiate the lesion from the surrounding skin. The speed function incorporates the chrominance properties of the lesion to stop evolution at the lesion boundary. Automatic initialization of the initial contour and chrominance-based speed function aids in providing robust and flexible segmentation. The proposed approach is tested on 200 images from PH2 and 900 images from ISBI 2016 datasets. Average accuracy, sensitivity, specificity, and overlap scores of 93.4, 87.6, 95.3, and 11.52% respectively are obtained for the PH2 dataset. Similarly, the proposed method resulted in average accuracy, sensitivity, specificity, and overlap scores of 94.6, 82.4, 97.2, and 7.20% respectively for the ISBI 2016 dataset. Statistical and quantitative analyses prove the reliability of the algorithm for incorporation in CAD systems. [Figure not available: see fulltext.].

AB - Automated segmentation and dermoscopic hair detection are one of the significant challenges in computer-aided diagnosis (CAD) of melanocytic lesions. Additionally, due to the presence of artifacts and variation in skin texture and smooth lesion boundaries, the accuracy of such methods gets hampered. The objective of this research is to develop an automated hair detection and lesion segmentation algorithm using lesion-specific properties to improve the accuracy. The aforementioned objective is achieved in two ways. Firstly, a novel hair detection algorithm is designed by considering the properties of dermoscopic hair. Second, a novel chroma-based geometric deformable model is used to effectively differentiate the lesion from the surrounding skin. The speed function incorporates the chrominance properties of the lesion to stop evolution at the lesion boundary. Automatic initialization of the initial contour and chrominance-based speed function aids in providing robust and flexible segmentation. The proposed approach is tested on 200 images from PH2 and 900 images from ISBI 2016 datasets. Average accuracy, sensitivity, specificity, and overlap scores of 93.4, 87.6, 95.3, and 11.52% respectively are obtained for the PH2 dataset. Similarly, the proposed method resulted in average accuracy, sensitivity, specificity, and overlap scores of 94.6, 82.4, 97.2, and 7.20% respectively for the ISBI 2016 dataset. Statistical and quantitative analyses prove the reliability of the algorithm for incorporation in CAD systems. [Figure not available: see fulltext.].

UR - http://www.scopus.com/inward/record.url?scp=85055127649&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85055127649&partnerID=8YFLogxK

U2 - 10.1007/s11517-018-1837-9

DO - 10.1007/s11517-018-1837-9

M3 - Article

AN - SCOPUS:85055127649

VL - 56

SP - 2051

EP - 2065

JO - Medical and Biological Engineering and Computing

JF - Medical and Biological Engineering and Computing

SN - 0140-0118

IS - 11

ER -