Development of a semi-automated combined PET and CT lung lesion segmentation framework

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

Abstract

Segmentation is one of the most important steps in automated medical diagnosis applications, which affects the accuracy of the overall system. In this paper, we propose a semi-automated segmentation method for extracting lung lesions from thoracic PET/CT images by combining low level processing and active contour techniques. The lesions are first segmented in PET images which are first converted to standardised uptake values (SUVs). The segmented PET images then serve as an initial contour for subsequent active contour segmentation of corresponding CT images. To evaluate its accuracy, the Jaccard Index (JI) was used as a measure of the accuracy of the segmented lesion compared to alternative segmentations from the QIN lung CT segmentation challenge, which is possible by registering the whole body PET/CT images to the corresponding thoracic CT images. The results show that our proposed technique has acceptable accuracy in lung lesion segmentation with JI values of around 0.8, especially when considering the variability of the alternative segmentations.

Original languageEnglish
Title of host publicationMedical Imaging 2017
Subtitle of host publicationBiomedical Applications in Molecular, Structural, and Functional Imaging
PublisherSPIE
Volume10137
ISBN (Electronic)9781510607194
DOIs
Publication statusPublished - 2017
EventMedical Imaging 2017: Biomedical Applications in Molecular, Structural, and Functional Imaging - Orlando, United States
Duration: 12 Feb 201714 Feb 2017

Other

OtherMedical Imaging 2017: Biomedical Applications in Molecular, Structural, and Functional Imaging
CountryUnited States
CityOrlando
Period12/2/1714/2/17

Fingerprint

lungs
lesions
Lung
Thorax
Processing

Keywords

  • Active Contour
  • Low level processing
  • Lung Lesion
  • PET/CT
  • Registration
  • Semi Automated
  • Thoracic CT
  • Volumetric Segmentation

ASJC Scopus subject areas

  • Atomic and Molecular Physics, and Optics
  • Electronic, Optical and Magnetic Materials
  • Biomaterials
  • Radiology Nuclear Medicine and imaging

Cite this

Rossi, F., Mokri, S. S., & Abd Rahni, A. A. (2017). Development of a semi-automated combined PET and CT lung lesion segmentation framework. In Medical Imaging 2017: Biomedical Applications in Molecular, Structural, and Functional Imaging (Vol. 10137). [101370B] SPIE. https://doi.org/10.1117/12.2256808

Development of a semi-automated combined PET and CT lung lesion segmentation framework. / Rossi, Farli; Mokri, Siti Salasiah; Abd Rahni, Ashrani Aizzuddin.

Medical Imaging 2017: Biomedical Applications in Molecular, Structural, and Functional Imaging. Vol. 10137 SPIE, 2017. 101370B.

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

Rossi, F, Mokri, SS & Abd Rahni, AA 2017, Development of a semi-automated combined PET and CT lung lesion segmentation framework. in Medical Imaging 2017: Biomedical Applications in Molecular, Structural, and Functional Imaging. vol. 10137, 101370B, SPIE, Medical Imaging 2017: Biomedical Applications in Molecular, Structural, and Functional Imaging, Orlando, United States, 12/2/17. https://doi.org/10.1117/12.2256808
Rossi F, Mokri SS, Abd Rahni AA. Development of a semi-automated combined PET and CT lung lesion segmentation framework. In Medical Imaging 2017: Biomedical Applications in Molecular, Structural, and Functional Imaging. Vol. 10137. SPIE. 2017. 101370B https://doi.org/10.1117/12.2256808
Rossi, Farli ; Mokri, Siti Salasiah ; Abd Rahni, Ashrani Aizzuddin. / Development of a semi-automated combined PET and CT lung lesion segmentation framework. Medical Imaging 2017: Biomedical Applications in Molecular, Structural, and Functional Imaging. Vol. 10137 SPIE, 2017.
@inproceedings{e8562e313c3d4671b9942dabd0aa3b69,
title = "Development of a semi-automated combined PET and CT lung lesion segmentation framework",
abstract = "Segmentation is one of the most important steps in automated medical diagnosis applications, which affects the accuracy of the overall system. In this paper, we propose a semi-automated segmentation method for extracting lung lesions from thoracic PET/CT images by combining low level processing and active contour techniques. The lesions are first segmented in PET images which are first converted to standardised uptake values (SUVs). The segmented PET images then serve as an initial contour for subsequent active contour segmentation of corresponding CT images. To evaluate its accuracy, the Jaccard Index (JI) was used as a measure of the accuracy of the segmented lesion compared to alternative segmentations from the QIN lung CT segmentation challenge, which is possible by registering the whole body PET/CT images to the corresponding thoracic CT images. The results show that our proposed technique has acceptable accuracy in lung lesion segmentation with JI values of around 0.8, especially when considering the variability of the alternative segmentations.",
keywords = "Active Contour, Low level processing, Lung Lesion, PET/CT, Registration, Semi Automated, Thoracic CT, Volumetric Segmentation",
author = "Farli Rossi and Mokri, {Siti Salasiah} and {Abd Rahni}, {Ashrani Aizzuddin}",
year = "2017",
doi = "10.1117/12.2256808",
language = "English",
volume = "10137",
booktitle = "Medical Imaging 2017",
publisher = "SPIE",

}

TY - GEN

T1 - Development of a semi-automated combined PET and CT lung lesion segmentation framework

AU - Rossi, Farli

AU - Mokri, Siti Salasiah

AU - Abd Rahni, Ashrani Aizzuddin

PY - 2017

Y1 - 2017

N2 - Segmentation is one of the most important steps in automated medical diagnosis applications, which affects the accuracy of the overall system. In this paper, we propose a semi-automated segmentation method for extracting lung lesions from thoracic PET/CT images by combining low level processing and active contour techniques. The lesions are first segmented in PET images which are first converted to standardised uptake values (SUVs). The segmented PET images then serve as an initial contour for subsequent active contour segmentation of corresponding CT images. To evaluate its accuracy, the Jaccard Index (JI) was used as a measure of the accuracy of the segmented lesion compared to alternative segmentations from the QIN lung CT segmentation challenge, which is possible by registering the whole body PET/CT images to the corresponding thoracic CT images. The results show that our proposed technique has acceptable accuracy in lung lesion segmentation with JI values of around 0.8, especially when considering the variability of the alternative segmentations.

AB - Segmentation is one of the most important steps in automated medical diagnosis applications, which affects the accuracy of the overall system. In this paper, we propose a semi-automated segmentation method for extracting lung lesions from thoracic PET/CT images by combining low level processing and active contour techniques. The lesions are first segmented in PET images which are first converted to standardised uptake values (SUVs). The segmented PET images then serve as an initial contour for subsequent active contour segmentation of corresponding CT images. To evaluate its accuracy, the Jaccard Index (JI) was used as a measure of the accuracy of the segmented lesion compared to alternative segmentations from the QIN lung CT segmentation challenge, which is possible by registering the whole body PET/CT images to the corresponding thoracic CT images. The results show that our proposed technique has acceptable accuracy in lung lesion segmentation with JI values of around 0.8, especially when considering the variability of the alternative segmentations.

KW - Active Contour

KW - Low level processing

KW - Lung Lesion

KW - PET/CT

KW - Registration

KW - Semi Automated

KW - Thoracic CT

KW - Volumetric Segmentation

UR - http://www.scopus.com/inward/record.url?scp=85020302953&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85020302953&partnerID=8YFLogxK

U2 - 10.1117/12.2256808

DO - 10.1117/12.2256808

M3 - Conference contribution

AN - SCOPUS:85020302953

VL - 10137

BT - Medical Imaging 2017

PB - SPIE

ER -