Automatic Lung Segmentation Using Control Feedback System: Morphology and Texture Paradigm

Norliza M. Noor, Joel C M Than, Omar M. Rijal, Rosminah M. Kassim, Ashari Yunus, Amir Zeki, Michele Anzidei, Luca Saba, Jasjit S. Suri

Research output: Contribution to journalArticle

25 Citations (Scopus)

Abstract

Interstitial Lung Disease (ILD) encompasses a wide array of diseases that share some common radiologic characteristics. When diagnosing such diseases, radiologists can be affected by heavy workload and fatigue thus decreasing diagnostic accuracy. Automatic segmentation is the first step in implementing a Computer Aided Diagnosis (CAD) that will help radiologists to improve diagnostic accuracy thereby reducing manual interpretation. Automatic segmentation proposed uses an initial thresholding and morphology based segmentation coupled with feedback that detects large deviations with a corrective segmentation. This feedback is analogous to a control system which allows detection of abnormal or severe lung disease and provides a feedback to an online segmentation improving the overall performance of the system. This feedback system encompasses a texture paradigm. In this study we studied 48 males and 48 female patients consisting of 15 normal and 81 abnormal patients. A senior radiologist chose the five levels needed for ILD diagnosis. The results of segmentation were displayed by showing the comparison of the automated and ground truth boundaries (courtesy of ImgTracer™ 1.0, AtheroPoint™ LLC, Roseville, CA, USA). The left lung’s performance of segmentation was 96.52 % for Jaccard Index and 98.21 % for Dice Similarity, 0.61 mm for Polyline Distance Metric (PDM), −1.15 % for Relative Area Error and 4.09 % Area Overlap Error. The right lung’s performance of segmentation was 97.24 % for Jaccard Index, 98.58 % for Dice Similarity, 0.61 mm for PDM, −0.03 % for Relative Area Error and 3.53 % for Area Overlap Error. The segmentation overall has an overall similarity of 98.4 %. The segmentation proposed is an accurate and fully automated system.

Original languageEnglish (US)
JournalJournal of Medical Systems
Volume39
Issue number3
DOIs
StatePublished - 2015

Fingerprint

Pulmonary diseases
Feedback control
Textures
Feedback
Lung
Interstitial Lung Diseases
Computer aided diagnosis
Workload
Lung Diseases
Fatigue
Fatigue of materials
Control systems
Radiologists

Keywords

  • High resolution computed tomography thorax (HRCT)
  • Interstitial lung disease
  • Lung segmentation

ASJC Scopus subject areas

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

Cite this

Automatic Lung Segmentation Using Control Feedback System : Morphology and Texture Paradigm. / Noor, Norliza M.; Than, Joel C M; Rijal, Omar M.; Kassim, Rosminah M.; Yunus, Ashari; Zeki, Amir; Anzidei, Michele; Saba, Luca; Suri, Jasjit S.

In: Journal of Medical Systems, Vol. 39, No. 3, 2015.

Research output: Contribution to journalArticle

Noor, NM, Than, JCM, Rijal, OM, Kassim, RM, Yunus, A, Zeki, A, Anzidei, M, Saba, L & Suri, JS 2015, 'Automatic Lung Segmentation Using Control Feedback System: Morphology and Texture Paradigm', Journal of Medical Systems, vol. 39, no. 3. https://doi.org/10.1007/s10916-015-0214-6
Noor, Norliza M. ; Than, Joel C M ; Rijal, Omar M. ; Kassim, Rosminah M. ; Yunus, Ashari ; Zeki, Amir ; Anzidei, Michele ; Saba, Luca ; Suri, Jasjit S. / Automatic Lung Segmentation Using Control Feedback System : Morphology and Texture Paradigm. In: Journal of Medical Systems. 2015 ; Vol. 39, No. 3.
@article{f02220aec4f34a929e47ddbba0ef32d9,
title = "Automatic Lung Segmentation Using Control Feedback System: Morphology and Texture Paradigm",
abstract = "Interstitial Lung Disease (ILD) encompasses a wide array of diseases that share some common radiologic characteristics. When diagnosing such diseases, radiologists can be affected by heavy workload and fatigue thus decreasing diagnostic accuracy. Automatic segmentation is the first step in implementing a Computer Aided Diagnosis (CAD) that will help radiologists to improve diagnostic accuracy thereby reducing manual interpretation. Automatic segmentation proposed uses an initial thresholding and morphology based segmentation coupled with feedback that detects large deviations with a corrective segmentation. This feedback is analogous to a control system which allows detection of abnormal or severe lung disease and provides a feedback to an online segmentation improving the overall performance of the system. This feedback system encompasses a texture paradigm. In this study we studied 48 males and 48 female patients consisting of 15 normal and 81 abnormal patients. A senior radiologist chose the five levels needed for ILD diagnosis. The results of segmentation were displayed by showing the comparison of the automated and ground truth boundaries (courtesy of ImgTracer™ 1.0, AtheroPoint™ LLC, Roseville, CA, USA). The left lung’s performance of segmentation was 96.52 {\%} for Jaccard Index and 98.21 {\%} for Dice Similarity, 0.61 mm for Polyline Distance Metric (PDM), −1.15 {\%} for Relative Area Error and 4.09 {\%} Area Overlap Error. The right lung’s performance of segmentation was 97.24 {\%} for Jaccard Index, 98.58 {\%} for Dice Similarity, 0.61 mm for PDM, −0.03 {\%} for Relative Area Error and 3.53 {\%} for Area Overlap Error. The segmentation overall has an overall similarity of 98.4 {\%}. The segmentation proposed is an accurate and fully automated system.",
keywords = "High resolution computed tomography thorax (HRCT), Interstitial lung disease, Lung segmentation",
author = "Noor, {Norliza M.} and Than, {Joel C M} and Rijal, {Omar M.} and Kassim, {Rosminah M.} and Ashari Yunus and Amir Zeki and Michele Anzidei and Luca Saba and Suri, {Jasjit S.}",
year = "2015",
doi = "10.1007/s10916-015-0214-6",
language = "English (US)",
volume = "39",
journal = "Journal of Medical Systems",
issn = "0148-5598",
publisher = "Springer New York",
number = "3",

}

TY - JOUR

T1 - Automatic Lung Segmentation Using Control Feedback System

T2 - Morphology and Texture Paradigm

AU - Noor, Norliza M.

AU - Than, Joel C M

AU - Rijal, Omar M.

AU - Kassim, Rosminah M.

AU - Yunus, Ashari

AU - Zeki, Amir

AU - Anzidei, Michele

AU - Saba, Luca

AU - Suri, Jasjit S.

PY - 2015

Y1 - 2015

N2 - Interstitial Lung Disease (ILD) encompasses a wide array of diseases that share some common radiologic characteristics. When diagnosing such diseases, radiologists can be affected by heavy workload and fatigue thus decreasing diagnostic accuracy. Automatic segmentation is the first step in implementing a Computer Aided Diagnosis (CAD) that will help radiologists to improve diagnostic accuracy thereby reducing manual interpretation. Automatic segmentation proposed uses an initial thresholding and morphology based segmentation coupled with feedback that detects large deviations with a corrective segmentation. This feedback is analogous to a control system which allows detection of abnormal or severe lung disease and provides a feedback to an online segmentation improving the overall performance of the system. This feedback system encompasses a texture paradigm. In this study we studied 48 males and 48 female patients consisting of 15 normal and 81 abnormal patients. A senior radiologist chose the five levels needed for ILD diagnosis. The results of segmentation were displayed by showing the comparison of the automated and ground truth boundaries (courtesy of ImgTracer™ 1.0, AtheroPoint™ LLC, Roseville, CA, USA). The left lung’s performance of segmentation was 96.52 % for Jaccard Index and 98.21 % for Dice Similarity, 0.61 mm for Polyline Distance Metric (PDM), −1.15 % for Relative Area Error and 4.09 % Area Overlap Error. The right lung’s performance of segmentation was 97.24 % for Jaccard Index, 98.58 % for Dice Similarity, 0.61 mm for PDM, −0.03 % for Relative Area Error and 3.53 % for Area Overlap Error. The segmentation overall has an overall similarity of 98.4 %. The segmentation proposed is an accurate and fully automated system.

AB - Interstitial Lung Disease (ILD) encompasses a wide array of diseases that share some common radiologic characteristics. When diagnosing such diseases, radiologists can be affected by heavy workload and fatigue thus decreasing diagnostic accuracy. Automatic segmentation is the first step in implementing a Computer Aided Diagnosis (CAD) that will help radiologists to improve diagnostic accuracy thereby reducing manual interpretation. Automatic segmentation proposed uses an initial thresholding and morphology based segmentation coupled with feedback that detects large deviations with a corrective segmentation. This feedback is analogous to a control system which allows detection of abnormal or severe lung disease and provides a feedback to an online segmentation improving the overall performance of the system. This feedback system encompasses a texture paradigm. In this study we studied 48 males and 48 female patients consisting of 15 normal and 81 abnormal patients. A senior radiologist chose the five levels needed for ILD diagnosis. The results of segmentation were displayed by showing the comparison of the automated and ground truth boundaries (courtesy of ImgTracer™ 1.0, AtheroPoint™ LLC, Roseville, CA, USA). The left lung’s performance of segmentation was 96.52 % for Jaccard Index and 98.21 % for Dice Similarity, 0.61 mm for Polyline Distance Metric (PDM), −1.15 % for Relative Area Error and 4.09 % Area Overlap Error. The right lung’s performance of segmentation was 97.24 % for Jaccard Index, 98.58 % for Dice Similarity, 0.61 mm for PDM, −0.03 % for Relative Area Error and 3.53 % for Area Overlap Error. The segmentation overall has an overall similarity of 98.4 %. The segmentation proposed is an accurate and fully automated system.

KW - High resolution computed tomography thorax (HRCT)

KW - Interstitial lung disease

KW - Lung segmentation

UR - http://www.scopus.com/inward/record.url?scp=84922469743&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84922469743&partnerID=8YFLogxK

U2 - 10.1007/s10916-015-0214-6

DO - 10.1007/s10916-015-0214-6

M3 - Article

C2 - 25666926

AN - SCOPUS:84922469743

VL - 39

JO - Journal of Medical Systems

JF - Journal of Medical Systems

SN - 0148-5598

IS - 3

ER -