Variational Bayes inference based segmentation of heterogeneous lymphoma volumes in dual-modality PET-CT images

Jiyong Wang; Yong Xia; Jiabin Wang; David Dagan Feng

doi:10.1109/DICTA.2011.52

Variational Bayes inference based segmentation of heterogeneous lymphoma volumes in dual-modality PET-CT images

Jiyong Wang
, Yong Xia
, Jiabin Wang
, David Dagan Feng

The University of Sydney
Royal Prince Alfred Hospital
Hong Kong Polytechnic University
Shanghai Jiao Tong University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Accurate segmentation of heterogeneous carcinoma lesions in medical images is vital to the treatment planning, assessment of therapy response and other oncological applications. With current state-of-the-art imaging modalities, the CT images enhance the interpretation of cancer functional abnormalities. We applied the variational Bayes inference (VBI) model on both anatomical and functional information for delineating lesion boundary. The model is improved by clinical meaningful initialisation. Clinical data consisting of eight lesions with inhomogeneous carcinoma distribution were used to evaluate the model accuracy. Our algorithm is capable of isolating lesions from background with higher accuracy comparing to the wildly used threshold (40% of SUVmax). The VBI segmentation error is less than 6.11% ± 4.92% which is much better than the results performed by fixed threshold method. The experimental results show that our novel statistic method can produce more accurate segmentation of heterogeneous lymphoma volume in PET-CT images.

Original language	English
Title of host publication	Proceedings - 2011 International Conference on Digital Image Computing
Subtitle of host publication	Techniques and Applications, DICTA 2011
Pages	274-278
Number of pages	5
DOIs	https://doi.org/10.1109/DICTA.2011.52
State	Published - 2011
Externally published	Yes
Event	2011 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2011 - Noosa, QLD, Australia Duration: 6 Dec 2011 → 8 Dec 2011

Publication series

Name	Proceedings - 2011 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2011

Conference

Conference	2011 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2011
Country/Territory	Australia
City	Noosa, QLD
Period	6/12/11 → 8/12/11

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being

Keywords

lymphoma
PET-CT
tumour segmentation
Variational Bayes Inference (VBI)

Access to Document

10.1109/DICTA.2011.52

Cite this

Wang, J., Xia, Y., Wang, J., & Feng, D. D. (2011). Variational Bayes inference based segmentation of heterogeneous lymphoma volumes in dual-modality PET-CT images. In Proceedings - 2011 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2011 (pp. 274-278). Article 6128694 (Proceedings - 2011 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2011). https://doi.org/10.1109/DICTA.2011.52

Wang, Jiyong ; Xia, Yong ; Wang, Jiabin et al. / Variational Bayes inference based segmentation of heterogeneous lymphoma volumes in dual-modality PET-CT images. Proceedings - 2011 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2011. 2011. pp. 274-278 (Proceedings - 2011 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2011).

@inproceedings{8419eab0c98a47b8b0d00bb85ee1497c,

title = "Variational Bayes inference based segmentation of heterogeneous lymphoma volumes in dual-modality PET-CT images",

abstract = "Accurate segmentation of heterogeneous carcinoma lesions in medical images is vital to the treatment planning, assessment of therapy response and other oncological applications. With current state-of-the-art imaging modalities, the CT images enhance the interpretation of cancer functional abnormalities. We applied the variational Bayes inference (VBI) model on both anatomical and functional information for delineating lesion boundary. The model is improved by clinical meaningful initialisation. Clinical data consisting of eight lesions with inhomogeneous carcinoma distribution were used to evaluate the model accuracy. Our algorithm is capable of isolating lesions from background with higher accuracy comparing to the wildly used threshold (40\% of SUVmax). The VBI segmentation error is less than 6.11\% ± 4.92\% which is much better than the results performed by fixed threshold method. The experimental results show that our novel statistic method can produce more accurate segmentation of heterogeneous lymphoma volume in PET-CT images.",

keywords = "lymphoma, PET-CT, tumour segmentation, Variational Bayes Inference (VBI)",

author = "Jiyong Wang and Yong Xia and Jiabin Wang and Feng, \{David Dagan\}",

year = "2011",

doi = "10.1109/DICTA.2011.52",

language = "英语",

isbn = "9780769545882",

series = "Proceedings - 2011 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2011",

pages = "274--278",

booktitle = "Proceedings - 2011 International Conference on Digital Image Computing",

note = "2011 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2011 ; Conference date: 06-12-2011 Through 08-12-2011",

}

Wang, J, Xia, Y, Wang, J & Feng, DD 2011, Variational Bayes inference based segmentation of heterogeneous lymphoma volumes in dual-modality PET-CT images. in Proceedings - 2011 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2011., 6128694, Proceedings - 2011 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2011, pp. 274-278, 2011 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2011, Noosa, QLD, Australia, 6/12/11. https://doi.org/10.1109/DICTA.2011.52

Variational Bayes inference based segmentation of heterogeneous lymphoma volumes in dual-modality PET-CT images. / Wang, Jiyong; Xia, Yong; Wang, Jiabin et al.
Proceedings - 2011 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2011. 2011. p. 274-278 6128694 (Proceedings - 2011 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2011).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Variational Bayes inference based segmentation of heterogeneous lymphoma volumes in dual-modality PET-CT images

AU - Wang, Jiyong

AU - Xia, Yong

AU - Wang, Jiabin

AU - Feng, David Dagan

PY - 2011

Y1 - 2011

N2 - Accurate segmentation of heterogeneous carcinoma lesions in medical images is vital to the treatment planning, assessment of therapy response and other oncological applications. With current state-of-the-art imaging modalities, the CT images enhance the interpretation of cancer functional abnormalities. We applied the variational Bayes inference (VBI) model on both anatomical and functional information for delineating lesion boundary. The model is improved by clinical meaningful initialisation. Clinical data consisting of eight lesions with inhomogeneous carcinoma distribution were used to evaluate the model accuracy. Our algorithm is capable of isolating lesions from background with higher accuracy comparing to the wildly used threshold (40% of SUVmax). The VBI segmentation error is less than 6.11% ± 4.92% which is much better than the results performed by fixed threshold method. The experimental results show that our novel statistic method can produce more accurate segmentation of heterogeneous lymphoma volume in PET-CT images.

AB - Accurate segmentation of heterogeneous carcinoma lesions in medical images is vital to the treatment planning, assessment of therapy response and other oncological applications. With current state-of-the-art imaging modalities, the CT images enhance the interpretation of cancer functional abnormalities. We applied the variational Bayes inference (VBI) model on both anatomical and functional information for delineating lesion boundary. The model is improved by clinical meaningful initialisation. Clinical data consisting of eight lesions with inhomogeneous carcinoma distribution were used to evaluate the model accuracy. Our algorithm is capable of isolating lesions from background with higher accuracy comparing to the wildly used threshold (40% of SUVmax). The VBI segmentation error is less than 6.11% ± 4.92% which is much better than the results performed by fixed threshold method. The experimental results show that our novel statistic method can produce more accurate segmentation of heterogeneous lymphoma volume in PET-CT images.

KW - lymphoma

KW - PET-CT

KW - tumour segmentation

KW - Variational Bayes Inference (VBI)

UR - https://www.scopus.com/pages/publications/84863026968

U2 - 10.1109/DICTA.2011.52

DO - 10.1109/DICTA.2011.52

M3 - 会议稿件

AN - SCOPUS:84863026968

SN - 9780769545882

T3 - Proceedings - 2011 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2011

SP - 274

EP - 278

BT - Proceedings - 2011 International Conference on Digital Image Computing

T2 - 2011 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2011

Y2 - 6 December 2011 through 8 December 2011

ER -

Wang J, Xia Y, Wang J, Feng DD. Variational Bayes inference based segmentation of heterogeneous lymphoma volumes in dual-modality PET-CT images. In Proceedings - 2011 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2011. 2011. p. 274-278. 6128694. (Proceedings - 2011 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2011). doi: 10.1109/DICTA.2011.52

Variational Bayes inference based segmentation of heterogeneous lymphoma volumes in dual-modality PET-CT images

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this