Automatic detection of larynx cancer from contrast-enhanced magnetic resonance images

Trushali Doshi, John Soraghan, Derek Grose, Kenneth MacKenzie, Lykourgos Petropoulakis

Research output: Chapter in Book/Report/Conference proceedingConference contribution book

Abstract

Detection of larynx cancer from medical imaging is important for the quantification and for the definition of target volumes in radiotherapy treatment planning (RTP). Magnetic resonance imaging (MRI) is being increasingly used in RTP due to its high resolution and excellent soft tissue contrast. Manually detecting larynx cancer from sequential MRI is time consuming and subjective. The large diversity of cancer in terms of geometry, non-distinct boundaries combined with the presence of normal anatomical regions close to the cancer regions necessitates the development of automatic and robust algorithms for this task. A new automatic algorithm for the detection of larynx cancer from 2D gadoliniumenhanced T1-weighted (T1+Gd) MRI to assist clinicians in RTP is presented. The algorithm employs edge detection using spatial neighborhood information of pixels and incorporates this information in a fuzzy c-means clustering process to robustly separate different tissues types. Furthermore, it utilizes the information of the expected cancerous location for cancer regions labeling. Comparison of this automatic detection system with manual clinical detection on real T1+Gd axial MRI slices of 2 patients (24 MRI slices) with visible larynx cancer yields an average dice similarity coefficient of 0.78±0.04 and average root mean square error of 1.82±0.28 mm. Preliminary results show that this fully automatic system can assist clinicians in RTP by obtaining quantifiable and non-subjective repeatable detection results in a particular time-efficient and unbiased fashion.

LanguageEnglish
Title of host publicationMedical Imaging 2015
Subtitle of host publicationComputer-Aided Diagnosis
EditorsLubomir M. Hadjiiski, Georgia D. Tourassi
Place of PublicationBellingham, WA.
Number of pages9
DOIs
Publication statusPublished - 8 May 2015
EventSPIE Medical Imaging Symposium 2015: Computer-Aided Diagnosis - Orlando, United States
Duration: 22 Feb 201525 Feb 2015

Publication series

NameProceedings of SPIE
Volume9414

Conference

ConferenceSPIE Medical Imaging Symposium 2015: Computer-Aided Diagnosis
CountryUnited States
CityOrlando
Period22/02/1525/02/15

Fingerprint

larynx
Laryngeal Neoplasms
Magnetic resonance
Magnetic resonance imaging
magnetic resonance
Radiotherapy
Magnetic Resonance Spectroscopy
cancer
Magnetic Resonance Imaging
Planning
planning
radiation therapy
Tissue
Neoplasms
Medical imaging
Edge detection
Diagnostic Imaging
Therapeutics
Mean square error
Labeling

Keywords

  • automatic detection
  • contrast-enhanced MRI
  • edge detection in spatial fuzzy c-means clustering
  • head and neck cancer
  • larynx cancer
  • radiotherapy treatment planning
  • throat detection

Cite this

Doshi, T., Soraghan, J., Grose, D., MacKenzie, K., & Petropoulakis, L. (2015). Automatic detection of larynx cancer from contrast-enhanced magnetic resonance images. In L. M. Hadjiiski, & G. D. Tourassi (Eds.), Medical Imaging 2015: Computer-Aided Diagnosis [94142N] (Proceedings of SPIE; Vol. 9414). Bellingham, WA.. https://doi.org/10.1117/12.2081864
Doshi, Trushali ; Soraghan, John ; Grose, Derek ; MacKenzie, Kenneth ; Petropoulakis, Lykourgos. / Automatic detection of larynx cancer from contrast-enhanced magnetic resonance images. Medical Imaging 2015: Computer-Aided Diagnosis. editor / Lubomir M. Hadjiiski ; Georgia D. Tourassi. Bellingham, WA., 2015. (Proceedings of SPIE).
@inproceedings{8fe4279bc49248bd8d42d2b0bd93ecc4,
title = "Automatic detection of larynx cancer from contrast-enhanced magnetic resonance images",
abstract = "Detection of larynx cancer from medical imaging is important for the quantification and for the definition of target volumes in radiotherapy treatment planning (RTP). Magnetic resonance imaging (MRI) is being increasingly used in RTP due to its high resolution and excellent soft tissue contrast. Manually detecting larynx cancer from sequential MRI is time consuming and subjective. The large diversity of cancer in terms of geometry, non-distinct boundaries combined with the presence of normal anatomical regions close to the cancer regions necessitates the development of automatic and robust algorithms for this task. A new automatic algorithm for the detection of larynx cancer from 2D gadoliniumenhanced T1-weighted (T1+Gd) MRI to assist clinicians in RTP is presented. The algorithm employs edge detection using spatial neighborhood information of pixels and incorporates this information in a fuzzy c-means clustering process to robustly separate different tissues types. Furthermore, it utilizes the information of the expected cancerous location for cancer regions labeling. Comparison of this automatic detection system with manual clinical detection on real T1+Gd axial MRI slices of 2 patients (24 MRI slices) with visible larynx cancer yields an average dice similarity coefficient of 0.78±0.04 and average root mean square error of 1.82±0.28 mm. Preliminary results show that this fully automatic system can assist clinicians in RTP by obtaining quantifiable and non-subjective repeatable detection results in a particular time-efficient and unbiased fashion.",
keywords = "automatic detection, contrast-enhanced MRI, edge detection in spatial fuzzy c-means clustering, head and neck cancer, larynx cancer, radiotherapy treatment planning, throat detection",
author = "Trushali Doshi and John Soraghan and Derek Grose and Kenneth MacKenzie and Lykourgos Petropoulakis",
year = "2015",
month = "5",
day = "8",
doi = "10.1117/12.2081864",
language = "English",
isbn = "9781628415049",
series = "Proceedings of SPIE",
editor = "Hadjiiski, {Lubomir M.} and Tourassi, {Georgia D.}",
booktitle = "Medical Imaging 2015",

}

Doshi, T, Soraghan, J, Grose, D, MacKenzie, K & Petropoulakis, L 2015, Automatic detection of larynx cancer from contrast-enhanced magnetic resonance images. in LM Hadjiiski & GD Tourassi (eds), Medical Imaging 2015: Computer-Aided Diagnosis., 94142N, Proceedings of SPIE, vol. 9414, Bellingham, WA., SPIE Medical Imaging Symposium 2015: Computer-Aided Diagnosis, Orlando, United States, 22/02/15. https://doi.org/10.1117/12.2081864

Automatic detection of larynx cancer from contrast-enhanced magnetic resonance images. / Doshi, Trushali; Soraghan, John; Grose, Derek; MacKenzie, Kenneth; Petropoulakis, Lykourgos.

Medical Imaging 2015: Computer-Aided Diagnosis. ed. / Lubomir M. Hadjiiski; Georgia D. Tourassi. Bellingham, WA., 2015. 94142N (Proceedings of SPIE; Vol. 9414).

Research output: Chapter in Book/Report/Conference proceedingConference contribution book

TY - GEN

T1 - Automatic detection of larynx cancer from contrast-enhanced magnetic resonance images

AU - Doshi, Trushali

AU - Soraghan, John

AU - Grose, Derek

AU - MacKenzie, Kenneth

AU - Petropoulakis, Lykourgos

PY - 2015/5/8

Y1 - 2015/5/8

N2 - Detection of larynx cancer from medical imaging is important for the quantification and for the definition of target volumes in radiotherapy treatment planning (RTP). Magnetic resonance imaging (MRI) is being increasingly used in RTP due to its high resolution and excellent soft tissue contrast. Manually detecting larynx cancer from sequential MRI is time consuming and subjective. The large diversity of cancer in terms of geometry, non-distinct boundaries combined with the presence of normal anatomical regions close to the cancer regions necessitates the development of automatic and robust algorithms for this task. A new automatic algorithm for the detection of larynx cancer from 2D gadoliniumenhanced T1-weighted (T1+Gd) MRI to assist clinicians in RTP is presented. The algorithm employs edge detection using spatial neighborhood information of pixels and incorporates this information in a fuzzy c-means clustering process to robustly separate different tissues types. Furthermore, it utilizes the information of the expected cancerous location for cancer regions labeling. Comparison of this automatic detection system with manual clinical detection on real T1+Gd axial MRI slices of 2 patients (24 MRI slices) with visible larynx cancer yields an average dice similarity coefficient of 0.78±0.04 and average root mean square error of 1.82±0.28 mm. Preliminary results show that this fully automatic system can assist clinicians in RTP by obtaining quantifiable and non-subjective repeatable detection results in a particular time-efficient and unbiased fashion.

AB - Detection of larynx cancer from medical imaging is important for the quantification and for the definition of target volumes in radiotherapy treatment planning (RTP). Magnetic resonance imaging (MRI) is being increasingly used in RTP due to its high resolution and excellent soft tissue contrast. Manually detecting larynx cancer from sequential MRI is time consuming and subjective. The large diversity of cancer in terms of geometry, non-distinct boundaries combined with the presence of normal anatomical regions close to the cancer regions necessitates the development of automatic and robust algorithms for this task. A new automatic algorithm for the detection of larynx cancer from 2D gadoliniumenhanced T1-weighted (T1+Gd) MRI to assist clinicians in RTP is presented. The algorithm employs edge detection using spatial neighborhood information of pixels and incorporates this information in a fuzzy c-means clustering process to robustly separate different tissues types. Furthermore, it utilizes the information of the expected cancerous location for cancer regions labeling. Comparison of this automatic detection system with manual clinical detection on real T1+Gd axial MRI slices of 2 patients (24 MRI slices) with visible larynx cancer yields an average dice similarity coefficient of 0.78±0.04 and average root mean square error of 1.82±0.28 mm. Preliminary results show that this fully automatic system can assist clinicians in RTP by obtaining quantifiable and non-subjective repeatable detection results in a particular time-efficient and unbiased fashion.

KW - automatic detection

KW - contrast-enhanced MRI

KW - edge detection in spatial fuzzy c-means clustering

KW - head and neck cancer

KW - larynx cancer

KW - radiotherapy treatment planning

KW - throat detection

UR - http://spie.org/Publications/Proceedings/Volume/9414

UR - http://www.scopus.com/inward/record.url?scp=84948807460&partnerID=8YFLogxK

U2 - 10.1117/12.2081864

DO - 10.1117/12.2081864

M3 - Conference contribution book

SN - 9781628415049

T3 - Proceedings of SPIE

BT - Medical Imaging 2015

A2 - Hadjiiski, Lubomir M.

A2 - Tourassi, Georgia D.

CY - Bellingham, WA.

ER -

Doshi T, Soraghan J, Grose D, MacKenzie K, Petropoulakis L. Automatic detection of larynx cancer from contrast-enhanced magnetic resonance images. In Hadjiiski LM, Tourassi GD, editors, Medical Imaging 2015: Computer-Aided Diagnosis. Bellingham, WA. 2015. 94142N. (Proceedings of SPIE). https://doi.org/10.1117/12.2081864