Unsupervised Tumour Segmentation in PET Based on Active Surface Modelling and Alpha Matting

Ziming Zeng; Reyer Zwiggelaar

Unsupervised Tumour Segmentation in PET Based on Active Surface Modelling and Alpha Matting

Ziming Zeng
, Reyer Zwiggelaar

Department of Computer Science

Shenyang Jianzhu University

Research output: Contribution to conference › Paper › peer-review

Abstract

This paper presents a novel, unsupervised method for segmenting tumours in PET data. The method uses region-based active surface modelling in a hierarchical scheme to eliminate segmentation errors, followed by an alpha matting step to further refine the segmentation. We have validated our method on real PET images of head-and-neck cancer patients as well as custom designed phantom PET images. Experiments show that our method can generate more accurate segmentation than some previous approaches.

Original language	English
Pages	131-136
Number of pages	6
Publication status	Published - 09 Jul 2012
Event	16th Conference on Medical Image Understanding and Analysis 2012 - Swansea, United Kingdom of Great Britain and Northern Ireland Duration: 09 Jul 2012 → 11 Jul 2012

Conference

Conference	16th Conference on Medical Image Understanding and Analysis 2012
Abbreviated title	MIUA 2012
Country/Territory	United Kingdom of Great Britain and Northern Ireland
City	Swansea
Period	09 Jul 2012 → 11 Jul 2012

UN SDGs

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

SDG 3 Good Health and Well-being

Permanent link

Persistent link

Cite this

@conference{1a9d93aa1de8437586135ceda2f274cd,

title = "Unsupervised Tumour Segmentation in PET Based on Active Surface Modelling and Alpha Matting",

abstract = "This paper presents a novel, unsupervised method for segmenting tumours in PET data. The method uses region-based active surface modelling in a hierarchical scheme to eliminate segmentation errors, followed by an alpha matting step to further refine the segmentation. We have validated our method on real PET images of head-and-neck cancer patients as well as custom designed phantom PET images. Experiments show that our method can generate more accurate segmentation than some previous approaches.",

author = "Ziming Zeng and Reyer Zwiggelaar",

year = "2012",

month = jul,

day = "9",

language = "English",

pages = "131--136",

note = "16th Conference on Medical Image Understanding and Analysis 2012, MIUA 2012 ; Conference date: 09-07-2012 Through 11-07-2012",

}

TY - CONF

T1 - Unsupervised Tumour Segmentation in PET Based on Active Surface Modelling and Alpha Matting

AU - Zeng, Ziming

AU - Zwiggelaar, Reyer

PY - 2012/7/9

Y1 - 2012/7/9

N2 - This paper presents a novel, unsupervised method for segmenting tumours in PET data. The method uses region-based active surface modelling in a hierarchical scheme to eliminate segmentation errors, followed by an alpha matting step to further refine the segmentation. We have validated our method on real PET images of head-and-neck cancer patients as well as custom designed phantom PET images. Experiments show that our method can generate more accurate segmentation than some previous approaches.

AB - This paper presents a novel, unsupervised method for segmenting tumours in PET data. The method uses region-based active surface modelling in a hierarchical scheme to eliminate segmentation errors, followed by an alpha matting step to further refine the segmentation. We have validated our method on real PET images of head-and-neck cancer patients as well as custom designed phantom PET images. Experiments show that our method can generate more accurate segmentation than some previous approaches.

M3 - Paper

SP - 131

EP - 136

T2 - 16th Conference on Medical Image Understanding and Analysis 2012

Y2 - 9 July 2012 through 11 July 2012

ER -

Unsupervised Tumour Segmentation in PET Based on Active Surface Modelling and Alpha Matting

Abstract

Conference

UN SDGs

Permanent link

Fingerprint

Cite this