A general framework for wireless capsule endoscopy study synopsis

Qian Zhao, Gerard Mullin, Max Q H Meng, Themistocles Dassopoulos, Rajesh Kumar

Research output: Contribution to journalArticle

Abstract

We present a general framework for analysis of wireless capsule endoscopy (CE) studies. The current available workstations provide a time-consuming and labor-intense work-flow for clinicians which requires the inspection of the full-length video. The development of a computer-aided diagnosis (CAD) CE workstation will have a great potential to reduce the diagnostic time and improve the accuracy of assessment. We propose a general framework based on hidden Markov models (HMMs) for study synopsis that forms the computational engine of our CAD workstation. Color, edge and texture features are first extracted and analyzed by a Support Vector Machine classifier, and then encoded as the observations for the HMM, uniquely combining the temporal information during the assessment. Experiments were performed on 13 full-length CE studies, instead of selected images previously reported. The results (e.g. 0.933 accuracy with 0.933 recall for detection of polyps) show that our framework achieved promising performance for multiple classification. We also report the patient-level CAD assessment of complete CE studies for multiple abnormalities, and the patient-level validation demonstrates the effectiveness and robustness of our methods.

Original languageEnglish (US)
Pages (from-to)108-116
Number of pages9
JournalComputerized Medical Imaging and Graphics
Volume41
DOIs
StatePublished - Apr 1 2015

Fingerprint

Capsule Endoscopy
Endoscopy
Computer aided diagnosis
Computer workstations
Hidden Markov models
Multiple Abnormalities
Workflow
Polyps
Support vector machines
Classifiers
Color
Textures
Inspection
Personnel
Engines
Experiments

Keywords

  • Computer-aided diagnosis
  • Hidden Markov models
  • Study synopsis
  • Supervised classification
  • Support vector machines
  • Wireless capsule endoscopy

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging
  • Health Informatics
  • Radiological and Ultrasound Technology
  • Computer Graphics and Computer-Aided Design
  • Computer Vision and Pattern Recognition
  • Medicine(all)

Cite this

A general framework for wireless capsule endoscopy study synopsis. / Zhao, Qian; Mullin, Gerard; Meng, Max Q H; Dassopoulos, Themistocles; Kumar, Rajesh.

In: Computerized Medical Imaging and Graphics, Vol. 41, 01.04.2015, p. 108-116.

Research output: Contribution to journalArticle

Zhao, Qian ; Mullin, Gerard ; Meng, Max Q H ; Dassopoulos, Themistocles ; Kumar, Rajesh. / A general framework for wireless capsule endoscopy study synopsis. In: Computerized Medical Imaging and Graphics. 2015 ; Vol. 41. pp. 108-116.
@article{92c24392707641dba26147d30f51c8e3,
title = "A general framework for wireless capsule endoscopy study synopsis",
abstract = "We present a general framework for analysis of wireless capsule endoscopy (CE) studies. The current available workstations provide a time-consuming and labor-intense work-flow for clinicians which requires the inspection of the full-length video. The development of a computer-aided diagnosis (CAD) CE workstation will have a great potential to reduce the diagnostic time and improve the accuracy of assessment. We propose a general framework based on hidden Markov models (HMMs) for study synopsis that forms the computational engine of our CAD workstation. Color, edge and texture features are first extracted and analyzed by a Support Vector Machine classifier, and then encoded as the observations for the HMM, uniquely combining the temporal information during the assessment. Experiments were performed on 13 full-length CE studies, instead of selected images previously reported. The results (e.g. 0.933 accuracy with 0.933 recall for detection of polyps) show that our framework achieved promising performance for multiple classification. We also report the patient-level CAD assessment of complete CE studies for multiple abnormalities, and the patient-level validation demonstrates the effectiveness and robustness of our methods.",
keywords = "Computer-aided diagnosis, Hidden Markov models, Study synopsis, Supervised classification, Support vector machines, Wireless capsule endoscopy",
author = "Qian Zhao and Gerard Mullin and Meng, {Max Q H} and Themistocles Dassopoulos and Rajesh Kumar",
year = "2015",
month = "4",
day = "1",
doi = "10.1016/j.compmedimag.2014.05.011",
language = "English (US)",
volume = "41",
pages = "108--116",
journal = "Computerized Medical Imaging and Graphics",
issn = "0895-6111",
publisher = "Elsevier Limited",

}

TY - JOUR

T1 - A general framework for wireless capsule endoscopy study synopsis

AU - Zhao, Qian

AU - Mullin, Gerard

AU - Meng, Max Q H

AU - Dassopoulos, Themistocles

AU - Kumar, Rajesh

PY - 2015/4/1

Y1 - 2015/4/1

N2 - We present a general framework for analysis of wireless capsule endoscopy (CE) studies. The current available workstations provide a time-consuming and labor-intense work-flow for clinicians which requires the inspection of the full-length video. The development of a computer-aided diagnosis (CAD) CE workstation will have a great potential to reduce the diagnostic time and improve the accuracy of assessment. We propose a general framework based on hidden Markov models (HMMs) for study synopsis that forms the computational engine of our CAD workstation. Color, edge and texture features are first extracted and analyzed by a Support Vector Machine classifier, and then encoded as the observations for the HMM, uniquely combining the temporal information during the assessment. Experiments were performed on 13 full-length CE studies, instead of selected images previously reported. The results (e.g. 0.933 accuracy with 0.933 recall for detection of polyps) show that our framework achieved promising performance for multiple classification. We also report the patient-level CAD assessment of complete CE studies for multiple abnormalities, and the patient-level validation demonstrates the effectiveness and robustness of our methods.

AB - We present a general framework for analysis of wireless capsule endoscopy (CE) studies. The current available workstations provide a time-consuming and labor-intense work-flow for clinicians which requires the inspection of the full-length video. The development of a computer-aided diagnosis (CAD) CE workstation will have a great potential to reduce the diagnostic time and improve the accuracy of assessment. We propose a general framework based on hidden Markov models (HMMs) for study synopsis that forms the computational engine of our CAD workstation. Color, edge and texture features are first extracted and analyzed by a Support Vector Machine classifier, and then encoded as the observations for the HMM, uniquely combining the temporal information during the assessment. Experiments were performed on 13 full-length CE studies, instead of selected images previously reported. The results (e.g. 0.933 accuracy with 0.933 recall for detection of polyps) show that our framework achieved promising performance for multiple classification. We also report the patient-level CAD assessment of complete CE studies for multiple abnormalities, and the patient-level validation demonstrates the effectiveness and robustness of our methods.

KW - Computer-aided diagnosis

KW - Hidden Markov models

KW - Study synopsis

KW - Supervised classification

KW - Support vector machines

KW - Wireless capsule endoscopy

UR - http://www.scopus.com/inward/record.url?scp=84923379792&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84923379792&partnerID=8YFLogxK

U2 - 10.1016/j.compmedimag.2014.05.011

DO - 10.1016/j.compmedimag.2014.05.011

M3 - Article

C2 - 24974010

AN - SCOPUS:84923379792

VL - 41

SP - 108

EP - 116

JO - Computerized Medical Imaging and Graphics

JF - Computerized Medical Imaging and Graphics

SN - 0895-6111

ER -