Automatic segmentation of clinical structures for RTP: Evaluation of a morphological approach

Gloria Bueno; Mark H. Fisher; Keith Burnham; John Mills; Oliver C. L. Hass

Automatic segmentation of clinical structures for RTP: Evaluation of a morphological approach

Gloria Bueno, Mark H. Fisher, Keith Burnham, John Mills, Oliver C. L. Hass

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

Abstract

This paper describes a morphological segmentation technique developed to assist clinicians and radiologists in conformal RTP. The system uses a watershed transformation to give an initial over-segmentation of a 2D CT; this is then refined by homotopy modification with markers identified automatically from the image histogram and the final segmentation is represented by a region adjacency graph which may be interactively edited by the user. To evaluate the system, 77 CT images from 11 patients have been segmented and the results compared against those produced by 5 clinical experts. The approach has enabled the quality of the automatic segmentation to be assessed in terms of the overall expert intra/inter variability. It has been proved good agreement between computer generated and expert manual delineations.

Original language	English
Pages	73-76
Number of pages	4
Publication status	Published - Jul 2001
Event	Medical Image Understanding and Analysis '01 - University of Birmingham, Birmingham, United Kingdom Duration: 16 Jul 2001 → 17 Jul 2001

Conference

Conference	Medical Image Understanding and Analysis '01
Abbreviated title	MIUA 2001
Country/Territory	United Kingdom
City	Birmingham
Period	16/07/01 → 17/07/01

Cite this

@conference{8d6210d047cf42f89d77a88afd4067f2,

title = "Automatic segmentation of clinical structures for RTP: Evaluation of a morphological approach",

abstract = "This paper describes a morphological segmentation technique developed to assist clinicians and radiologists in conformal RTP. The system uses a watershed transformation to give an initial over-segmentation of a 2D CT; this is then refined by homotopy modification with markers identified automatically from the image histogram and the final segmentation is represented by a region adjacency graph which may be interactively edited by the user. To evaluate the system, 77 CT images from 11 patients have been segmented and the results compared against those produced by 5 clinical experts. The approach has enabled the quality of the automatic segmentation to be assessed in terms of the overall expert intra/inter variability. It has been proved good agreement between computer generated and expert manual delineations.",

author = "Gloria Bueno and Fisher, {Mark H.} and Keith Burnham and John Mills and Hass, {Oliver C. L.}",

year = "2001",

month = jul,

language = "English",

pages = "73--76",

note = "Medical Image Understanding and Analysis '01, MIUA 2001 ; Conference date: 16-07-2001 Through 17-07-2001",

}

TY - CONF

T1 - Automatic segmentation of clinical structures for RTP: Evaluation of a morphological approach

AU - Bueno, Gloria

AU - Fisher, Mark H.

AU - Burnham, Keith

AU - Mills, John

AU - Hass, Oliver C. L.

PY - 2001/7

Y1 - 2001/7

N2 - This paper describes a morphological segmentation technique developed to assist clinicians and radiologists in conformal RTP. The system uses a watershed transformation to give an initial over-segmentation of a 2D CT; this is then refined by homotopy modification with markers identified automatically from the image histogram and the final segmentation is represented by a region adjacency graph which may be interactively edited by the user. To evaluate the system, 77 CT images from 11 patients have been segmented and the results compared against those produced by 5 clinical experts. The approach has enabled the quality of the automatic segmentation to be assessed in terms of the overall expert intra/inter variability. It has been proved good agreement between computer generated and expert manual delineations.

AB - This paper describes a morphological segmentation technique developed to assist clinicians and radiologists in conformal RTP. The system uses a watershed transformation to give an initial over-segmentation of a 2D CT; this is then refined by homotopy modification with markers identified automatically from the image histogram and the final segmentation is represented by a region adjacency graph which may be interactively edited by the user. To evaluate the system, 77 CT images from 11 patients have been segmented and the results compared against those produced by 5 clinical experts. The approach has enabled the quality of the automatic segmentation to be assessed in terms of the overall expert intra/inter variability. It has been proved good agreement between computer generated and expert manual delineations.

M3 - Paper

SP - 73

EP - 76

T2 - Medical Image Understanding and Analysis '01

Y2 - 16 July 2001 through 17 July 2001

ER -