Application of the ordered logit model to optimising Frangi filter parameters for segmentation of perivascular spaces

Lucia Ballerini; R Lovreglio; Maria Valdes Hernandez; Victor Gonzalez-Castro; Susana Muñoz Maniega; Enrico Pellegrini; Mark Bastin; Ian Deary; Joanna Wardlaw

doi:10.1016/j.procs.2016.07.011

Application of the ordered logit model to optimising Frangi filter parameters for segmentation of perivascular spaces

Lucia Ballerini, R Lovreglio, Maria Valdes Hernandez, Victor Gonzalez-Castro, Susana Muñoz Maniega, Enrico Pellegrini, Mark Bastin, Ian Deary, Joanna Wardlaw

Research output: Contribution to journal › Article › peer-review

Abstract / Description of output

Segmentation of perivascular spaces (PVS) from brain magnetic resonance images (MRI) is important for understanding the brain's lymphatic system and its relationship with neurological diseases. The Frangi filter might be a valuable tool for this purpose. However, its parameters need to be adjusted in response to the variability in the scanner's parameters and study protocols. Knowing the neuroradiological ratings of the PVS, we used the ordered logit model to optimise Frangi filter parameters. The PVS volume obtained significantly and strongly correlated with neuroradiological assessments (Spearman's ρ=0.75, p < 0.001), suggesting that the ordered logit model could be a good alternative to conventional optimisation frameworks for segmenting PVS on MRI.

Original language	English
Pages (from-to)	61-67
Journal	Procedia Computer Science
DOIs	https://doi.org/10.1016/j.procs.2016.07.011
Publication status	Published - 25 Jul 2016

Access to Document

10.1016/j.procs.2016.07.011

Application of the Ordered Logit Model to Optimising Frangi Filter Parameters for Segmentation of Perivascular SpacesAccepted author manuscript, 910 KB
Application of the Ordered Logit Model to Optimising Frangi Filter Parameters for Segmentation of Perivascular Spaces
© 2016 The Author(s). Published by Elsevier B.V.
Final published version, 307 KB

http://www.sciencedirect.com/science/article/pii/S1877050916311899

4 Finished

Row Fogo Charitable Trust Funds
Wardlaw, J.
UK-based charities
1/10/15 → 30/09/18
Project: Research
Brain imaging and cognitive ageing in the Lothian Birth Cohort 1936: III
Wardlaw, J., Bastin, M. & Deary, I.
MRC
1/05/15 → 30/04/19
Project: Research
RA2661 Centre for Cognitive Ageing and Cognitive Epidemiology Phase 2. Main Budget.
Deary, I., Gale, C., Holmes, M., Logie, P., Maclullich, A., Porteous, D., Seckl, J., Starr, J., Wardlaw, J. & Okely, J.
1/09/13 → 31/08/19
Project: Research

1 Poster

Quantitative Measurements of Enlarged Perivascular Spaces in the Brain are Associated with Retinal Microvascular Parameters
Ballerini, L., McGrory, S., Lovreglio, R., Valdes Hernandez, M., Pellegrini, E., MacGillivray, T., Trucco, E., Muñoz Maniega, S., Henderson, R., Taylor, A., Bastin, M., Deary, I. & Wardlaw, J., 11 May 2017.
Research output: Contribution to conference › Poster › peer-review

2 Participation in conference
1 Participation in workshop, seminar, course

Symposium on Small Vessel Disease of the Brain
Lucia Ballerini (Participant)
11 May 2017
Activity: Participating in or organising an event types › Participation in workshop, seminar, course
Medical Image Understanding and Analysis (MIUA 2016)
Lucia Ballerini (Speaker)
7 Jul 2016
Activity: Participating in or organising an event types › Participation in conference
CCACE Annual Research Day 2016
Lucia Ballerini (Speaker)
6 Sept 2016
Activity: Participating in or organising an event types › Participation in conference

Cite this

@article{6a45f12ef7074bd8aff5027653c34425,

title = "Application of the ordered logit model to optimising Frangi filter parameters for segmentation of perivascular spaces",

abstract = "Segmentation of perivascular spaces (PVS) from brain magnetic resonance images (MRI) is important for understanding the brain's lymphatic system and its relationship with neurological diseases. The Frangi filter might be a valuable tool for this purpose. However, its parameters need to be adjusted in response to the variability in the scanner's parameters and study protocols. Knowing the neuroradiological ratings of the PVS, we used the ordered logit model to optimise Frangi filter parameters. The PVS volume obtained significantly and strongly correlated with neuroradiological assessments (Spearman's ρ=0.75, p < 0.001), suggesting that the ordered logit model could be a good alternative to conventional optimisation frameworks for segmenting PVS on MRI.",

author = "Lucia Ballerini and R Lovreglio and {Valdes Hernandez}, Maria and Victor Gonzalez-Castro and {Mu{\~n}oz Maniega}, Susana and Enrico Pellegrini and Mark Bastin and Ian Deary and Joanna Wardlaw",

year = "2016",

month = jul,

day = "25",

doi = "10.1016/j.procs.2016.07.011",

language = "English",

pages = "61--67",

journal = "Procedia Computer Science",

issn = "1877-0509",

publisher = "Elsevier BV",

}

TY - JOUR

T1 - Application of the ordered logit model to optimising Frangi filter parameters for segmentation of perivascular spaces

AU - Ballerini, Lucia

AU - Lovreglio, R

AU - Valdes Hernandez, Maria

AU - Gonzalez-Castro, Victor

AU - Muñoz Maniega, Susana

AU - Pellegrini, Enrico

AU - Bastin, Mark

AU - Deary, Ian

AU - Wardlaw, Joanna

PY - 2016/7/25

Y1 - 2016/7/25

N2 - Segmentation of perivascular spaces (PVS) from brain magnetic resonance images (MRI) is important for understanding the brain's lymphatic system and its relationship with neurological diseases. The Frangi filter might be a valuable tool for this purpose. However, its parameters need to be adjusted in response to the variability in the scanner's parameters and study protocols. Knowing the neuroradiological ratings of the PVS, we used the ordered logit model to optimise Frangi filter parameters. The PVS volume obtained significantly and strongly correlated with neuroradiological assessments (Spearman's ρ=0.75, p < 0.001), suggesting that the ordered logit model could be a good alternative to conventional optimisation frameworks for segmenting PVS on MRI.

AB - Segmentation of perivascular spaces (PVS) from brain magnetic resonance images (MRI) is important for understanding the brain's lymphatic system and its relationship with neurological diseases. The Frangi filter might be a valuable tool for this purpose. However, its parameters need to be adjusted in response to the variability in the scanner's parameters and study protocols. Knowing the neuroradiological ratings of the PVS, we used the ordered logit model to optimise Frangi filter parameters. The PVS volume obtained significantly and strongly correlated with neuroradiological assessments (Spearman's ρ=0.75, p < 0.001), suggesting that the ordered logit model could be a good alternative to conventional optimisation frameworks for segmenting PVS on MRI.

U2 - 10.1016/j.procs.2016.07.011

DO - 10.1016/j.procs.2016.07.011

M3 - Article

SN - 1877-0509

SP - 61

EP - 67

JO - Procedia Computer Science

JF - Procedia Computer Science

ER -

Application of the ordered logit model to optimising Frangi filter parameters for segmentation of perivascular spaces

Abstract / Description of output

Access to Document

Fingerprint

Projects

Row Fogo Charitable Trust Funds

Brain imaging and cognitive ageing in the Lothian Birth Cohort 1936: III

RA2661 Centre for Cognitive Ageing and Cognitive Epidemiology Phase 2. Main Budget.

Research output

Quantitative Measurements of Enlarged Perivascular Spaces in the Brain are Associated with Retinal Microvascular Parameters

Activities

Symposium on Small Vessel Disease of the Brain

Medical Image Understanding and Analysis (MIUA 2016)

CCACE Annual Research Day 2016

Cite this