A multiscale decomposition approach to detect abnormal vasculature in the optic disc

Carla Agurto; Honggang Yu; Victor Murray; Marios S. Pattichis; Sheila Nemeth; Simon Barriga; Peter Soliz

doi:10.1016/j.compmedimag.2015.01.001

A multiscale decomposition approach to detect abnormal vasculature in the optic disc

Carla Agurto, Honggang Yu, Victor Murray, Marios S. Pattichis, Sheila Nemeth, Simon Barriga, Peter Soliz

Producción científica: Contribución a una revista › Artículo › revisión exhaustiva

17 Citas (Scopus)

Resumen

This paper presents a multiscale method to detect neovascularization in the optic disc (NVD) using fundus images. Our method is applied to a manually selected region of interest (ROI) containing the optic disc. All the vessels in the ROI are segmented by adaptively combining contrast enhancement methods with a vessel segmentation technique. Textural features extracted using multiscale amplitude-modulation frequency-modulation, morphological granulometry, and fractal dimension are used. A linear SVM is used to perform the classification, which is tested by means of 10-fold cross-validation. The performance is evaluated using 300 images achieving an AUC of 0.93 with maximum accuracy of 88%.

Idioma original	Inglés
Páginas (desde-hasta)	137-149
Número de páginas	13
Publicación	Computerized Medical Imaging and Graphics
Volumen	43
DOI	https://doi.org/10.1016/j.compmedimag.2015.01.001
Estado	Publicada - 1 jul. 2015
Publicado de forma externa	Sí

ODS de las Naciones Unidas

Este resultado contribuye a los siguientes Objetivos de Desarrollo Sostenible

Acceder al documento

10.1016/j.compmedimag.2015.01.001

Otros archivos y enlaces

Enlace a la publicación en Scopus

Citar esto

@article{18ee8216d24340d6a9816e751f61bfbd,

title = "A multiscale decomposition approach to detect abnormal vasculature in the optic disc",

abstract = "This paper presents a multiscale method to detect neovascularization in the optic disc (NVD) using fundus images. Our method is applied to a manually selected region of interest (ROI) containing the optic disc. All the vessels in the ROI are segmented by adaptively combining contrast enhancement methods with a vessel segmentation technique. Textural features extracted using multiscale amplitude-modulation frequency-modulation, morphological granulometry, and fractal dimension are used. A linear SVM is used to perform the classification, which is tested by means of 10-fold cross-validation. The performance is evaluated using 300 images achieving an AUC of 0.93 with maximum accuracy of 88%.",

keywords = "Amplitude-modulation frequency-modulation, Diabetic retinopathy, Fractal dimension, Granulometry, Neovascularization",

author = "Carla Agurto and Honggang Yu and Victor Murray and Pattichis, {Marios S.} and Sheila Nemeth and Simon Barriga and Peter Soliz",

note = "Publisher Copyright: {\textcopyright} 2015 Elsevier Ltd.",

year = "2015",

month = jul,

day = "1",

doi = "10.1016/j.compmedimag.2015.01.001",

language = "English",

volume = "43",

pages = "137--149",

journal = "Computerized Medical Imaging and Graphics",

issn = "0895-6111",

}

TY - JOUR

T1 - A multiscale decomposition approach to detect abnormal vasculature in the optic disc

AU - Agurto, Carla

AU - Yu, Honggang

AU - Murray, Victor

AU - Pattichis, Marios S.

AU - Nemeth, Sheila

AU - Barriga, Simon

AU - Soliz, Peter

PY - 2015/7/1

Y1 - 2015/7/1

N2 - This paper presents a multiscale method to detect neovascularization in the optic disc (NVD) using fundus images. Our method is applied to a manually selected region of interest (ROI) containing the optic disc. All the vessels in the ROI are segmented by adaptively combining contrast enhancement methods with a vessel segmentation technique. Textural features extracted using multiscale amplitude-modulation frequency-modulation, morphological granulometry, and fractal dimension are used. A linear SVM is used to perform the classification, which is tested by means of 10-fold cross-validation. The performance is evaluated using 300 images achieving an AUC of 0.93 with maximum accuracy of 88%.

AB - This paper presents a multiscale method to detect neovascularization in the optic disc (NVD) using fundus images. Our method is applied to a manually selected region of interest (ROI) containing the optic disc. All the vessels in the ROI are segmented by adaptively combining contrast enhancement methods with a vessel segmentation technique. Textural features extracted using multiscale amplitude-modulation frequency-modulation, morphological granulometry, and fractal dimension are used. A linear SVM is used to perform the classification, which is tested by means of 10-fold cross-validation. The performance is evaluated using 300 images achieving an AUC of 0.93 with maximum accuracy of 88%.

KW - Amplitude-modulation frequency-modulation

KW - Diabetic retinopathy

KW - Fractal dimension

KW - Granulometry

KW - Neovascularization

UR - http://www.scopus.com/inward/record.url?scp=84929959982&partnerID=8YFLogxK

U2 - 10.1016/j.compmedimag.2015.01.001

DO - 10.1016/j.compmedimag.2015.01.001

M3 - Article

C2 - 25698545

AN - SCOPUS:84929959982

SN - 0895-6111

VL - 43

SP - 137

EP - 149

JO - Computerized Medical Imaging and Graphics

JF - Computerized Medical Imaging and Graphics

ER -

A multiscale decomposition approach to detect abnormal vasculature in the optic disc

Resumen

ODS de las Naciones Unidas

Acceder al documento

Otros archivos y enlaces

Huella

Citar esto