Incorporating domain knowledge into the fuzzy connectedness framework

Application to brain lesion volume estimation in multiple sclerosis

Mark A. Horsfield, Rohit Bakshi, Marco Rovaris, Mara A. Rocca, Venkata S R Dandamudi, Paola Valsasina, Elda Judica, Fulvio Lucchini, Charles R G Guttmann, Maria Pia Sormani, Massimo Filippi

Research output: Contribution to journalArticle

22 Citations (Scopus)

Abstract

A method for incorporating prior knowledge into the fuzzy connectedness image segmentation framework is presented. This prior knowledge is in the form of probabilistic feature distribution and feature size maps, in a standard anatomical space, and "intensity hints" selected by the user that allow for a skewed distribution of the feature intensity characteristics. The fuzzy affinity between pixels is modified to encapsulate this domain knowledge. The method was tested by using it to segment brain lesions in patients with multiple sclerosis, and the results compared to an established method for lesion outlining based on edge detection and contour following. With the fuzzy connections (FC) method, the user is required to identify each lesion with a mouse click, to provide a set of seed pixels. The algorithm then grows the features from, the seeds to define the lesions as a set of objects with fuzzy connectedness above a pre-set threshold. The FC method gave improved inter-observer reproducibility of lesion volumes, and the set of pixels determined to be lesion was more consistent compared to the contouring method. The operator interaction time required to evaluate one subject was reduced from an average of 111 minutes with contouring to 16 minutes with the FC method.

Original languageEnglish (US)
Pages (from-to)1670-1680
Number of pages11
JournalIEEE Transactions on Medical Imaging
Volume26
Issue number12
DOIs
StatePublished - Dec 2007
Externally publishedYes

Fingerprint

Multiple Sclerosis
Brain
Pixels
Seed
Edge detection
Image segmentation
Seeds

Keywords

  • Fuzzy connectedness
  • Lesion volume
  • MRI
  • Multiple sclerosis
  • Prior knowledge

ASJC Scopus subject areas

  • Biomedical Engineering
  • Radiology Nuclear Medicine and imaging
  • Radiological and Ultrasound Technology
  • Electrical and Electronic Engineering
  • Computer Science Applications
  • Computational Theory and Mathematics

Cite this

Incorporating domain knowledge into the fuzzy connectedness framework : Application to brain lesion volume estimation in multiple sclerosis. / Horsfield, Mark A.; Bakshi, Rohit; Rovaris, Marco; Rocca, Mara A.; Dandamudi, Venkata S R; Valsasina, Paola; Judica, Elda; Lucchini, Fulvio; Guttmann, Charles R G; Sormani, Maria Pia; Filippi, Massimo.

In: IEEE Transactions on Medical Imaging, Vol. 26, No. 12, 12.2007, p. 1670-1680.

Research output: Contribution to journalArticle

Horsfield, MA, Bakshi, R, Rovaris, M, Rocca, MA, Dandamudi, VSR, Valsasina, P, Judica, E, Lucchini, F, Guttmann, CRG, Sormani, MP & Filippi, M 2007, 'Incorporating domain knowledge into the fuzzy connectedness framework: Application to brain lesion volume estimation in multiple sclerosis', IEEE Transactions on Medical Imaging, vol. 26, no. 12, pp. 1670-1680. https://doi.org/10.1109/TMI.2007.901431
Horsfield, Mark A. ; Bakshi, Rohit ; Rovaris, Marco ; Rocca, Mara A. ; Dandamudi, Venkata S R ; Valsasina, Paola ; Judica, Elda ; Lucchini, Fulvio ; Guttmann, Charles R G ; Sormani, Maria Pia ; Filippi, Massimo. / Incorporating domain knowledge into the fuzzy connectedness framework : Application to brain lesion volume estimation in multiple sclerosis. In: IEEE Transactions on Medical Imaging. 2007 ; Vol. 26, No. 12. pp. 1670-1680.
@article{2f24519a111f4600b6feb662e1fc709f,
title = "Incorporating domain knowledge into the fuzzy connectedness framework: Application to brain lesion volume estimation in multiple sclerosis",
abstract = "A method for incorporating prior knowledge into the fuzzy connectedness image segmentation framework is presented. This prior knowledge is in the form of probabilistic feature distribution and feature size maps, in a standard anatomical space, and {"}intensity hints{"} selected by the user that allow for a skewed distribution of the feature intensity characteristics. The fuzzy affinity between pixels is modified to encapsulate this domain knowledge. The method was tested by using it to segment brain lesions in patients with multiple sclerosis, and the results compared to an established method for lesion outlining based on edge detection and contour following. With the fuzzy connections (FC) method, the user is required to identify each lesion with a mouse click, to provide a set of seed pixels. The algorithm then grows the features from, the seeds to define the lesions as a set of objects with fuzzy connectedness above a pre-set threshold. The FC method gave improved inter-observer reproducibility of lesion volumes, and the set of pixels determined to be lesion was more consistent compared to the contouring method. The operator interaction time required to evaluate one subject was reduced from an average of 111 minutes with contouring to 16 minutes with the FC method.",
keywords = "Fuzzy connectedness, Lesion volume, MRI, Multiple sclerosis, Prior knowledge",
author = "Horsfield, {Mark A.} and Rohit Bakshi and Marco Rovaris and Rocca, {Mara A.} and Dandamudi, {Venkata S R} and Paola Valsasina and Elda Judica and Fulvio Lucchini and Guttmann, {Charles R G} and Sormani, {Maria Pia} and Massimo Filippi",
year = "2007",
month = "12",
doi = "10.1109/TMI.2007.901431",
language = "English (US)",
volume = "26",
pages = "1670--1680",
journal = "IEEE Transactions on Medical Imaging",
issn = "0278-0062",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "12",

}

TY - JOUR

T1 - Incorporating domain knowledge into the fuzzy connectedness framework

T2 - Application to brain lesion volume estimation in multiple sclerosis

AU - Horsfield, Mark A.

AU - Bakshi, Rohit

AU - Rovaris, Marco

AU - Rocca, Mara A.

AU - Dandamudi, Venkata S R

AU - Valsasina, Paola

AU - Judica, Elda

AU - Lucchini, Fulvio

AU - Guttmann, Charles R G

AU - Sormani, Maria Pia

AU - Filippi, Massimo

PY - 2007/12

Y1 - 2007/12

N2 - A method for incorporating prior knowledge into the fuzzy connectedness image segmentation framework is presented. This prior knowledge is in the form of probabilistic feature distribution and feature size maps, in a standard anatomical space, and "intensity hints" selected by the user that allow for a skewed distribution of the feature intensity characteristics. The fuzzy affinity between pixels is modified to encapsulate this domain knowledge. The method was tested by using it to segment brain lesions in patients with multiple sclerosis, and the results compared to an established method for lesion outlining based on edge detection and contour following. With the fuzzy connections (FC) method, the user is required to identify each lesion with a mouse click, to provide a set of seed pixels. The algorithm then grows the features from, the seeds to define the lesions as a set of objects with fuzzy connectedness above a pre-set threshold. The FC method gave improved inter-observer reproducibility of lesion volumes, and the set of pixels determined to be lesion was more consistent compared to the contouring method. The operator interaction time required to evaluate one subject was reduced from an average of 111 minutes with contouring to 16 minutes with the FC method.

AB - A method for incorporating prior knowledge into the fuzzy connectedness image segmentation framework is presented. This prior knowledge is in the form of probabilistic feature distribution and feature size maps, in a standard anatomical space, and "intensity hints" selected by the user that allow for a skewed distribution of the feature intensity characteristics. The fuzzy affinity between pixels is modified to encapsulate this domain knowledge. The method was tested by using it to segment brain lesions in patients with multiple sclerosis, and the results compared to an established method for lesion outlining based on edge detection and contour following. With the fuzzy connections (FC) method, the user is required to identify each lesion with a mouse click, to provide a set of seed pixels. The algorithm then grows the features from, the seeds to define the lesions as a set of objects with fuzzy connectedness above a pre-set threshold. The FC method gave improved inter-observer reproducibility of lesion volumes, and the set of pixels determined to be lesion was more consistent compared to the contouring method. The operator interaction time required to evaluate one subject was reduced from an average of 111 minutes with contouring to 16 minutes with the FC method.

KW - Fuzzy connectedness

KW - Lesion volume

KW - MRI

KW - Multiple sclerosis

KW - Prior knowledge

UR - http://www.scopus.com/inward/record.url?scp=36549088081&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=36549088081&partnerID=8YFLogxK

U2 - 10.1109/TMI.2007.901431

DO - 10.1109/TMI.2007.901431

M3 - Article

VL - 26

SP - 1670

EP - 1680

JO - IEEE Transactions on Medical Imaging

JF - IEEE Transactions on Medical Imaging

SN - 0278-0062

IS - 12

ER -