Statistical atlas based exudate segmentation

Sharib Ali, Désiré Sidibé, Kedir M. Adal, Luca Giancardo, Edward Chaum, Thomas P. Karnowski, Fabrice Mériaudeau

Research output: Contribution to journalArticle

34 Citations (Scopus)

Abstract

Diabetic macular edema (DME) is characterized by hard exudates. In this article, we propose a novel statistical atlas based method for segmentation of such exudates. Any test fundus image is first warped on the atlas co-ordinate and then a distance map is obtained with the mean atlas image. This leaves behind the candidate lesions. Post-processing schemes are introduced for final segmentation of the exudate. Experiments with the publicly available HEI-MED data-set shows good performance of the method. A lesion localization fraction of 82.5% at 35% of non-lesion localization fraction on the FROC curve is obtained. The method is also compared to few most recent reference methods.

Original languageEnglish (US)
Pages (from-to)358-368
Number of pages11
JournalComputerized Medical Imaging and Graphics
Volume37
Issue number5-6
DOIs
StatePublished - Jul 1 2013

Fingerprint

Atlases
Exudates and Transudates
Processing
Experiments
Macular Edema

All Science Journal Classification (ASJC) codes

  • Radiological and Ultrasound Technology
  • Radiology Nuclear Medicine and imaging
  • Computer Vision and Pattern Recognition
  • Health Informatics
  • Computer Graphics and Computer-Aided Design

Cite this

Ali, S., Sidibé, D., Adal, K. M., Giancardo, L., Chaum, E., Karnowski, T. P., & Mériaudeau, F. (2013). Statistical atlas based exudate segmentation. Computerized Medical Imaging and Graphics, 37(5-6), 358-368. https://doi.org/10.1016/j.compmedimag.2013.06.006

Statistical atlas based exudate segmentation. / Ali, Sharib; Sidibé, Désiré; Adal, Kedir M.; Giancardo, Luca; Chaum, Edward; Karnowski, Thomas P.; Mériaudeau, Fabrice.

In: Computerized Medical Imaging and Graphics, Vol. 37, No. 5-6, 01.07.2013, p. 358-368.

Research output: Contribution to journalArticle

Ali, S, Sidibé, D, Adal, KM, Giancardo, L, Chaum, E, Karnowski, TP & Mériaudeau, F 2013, 'Statistical atlas based exudate segmentation', Computerized Medical Imaging and Graphics, vol. 37, no. 5-6, pp. 358-368. https://doi.org/10.1016/j.compmedimag.2013.06.006
Ali S, Sidibé D, Adal KM, Giancardo L, Chaum E, Karnowski TP et al. Statistical atlas based exudate segmentation. Computerized Medical Imaging and Graphics. 2013 Jul 1;37(5-6):358-368. https://doi.org/10.1016/j.compmedimag.2013.06.006
Ali, Sharib ; Sidibé, Désiré ; Adal, Kedir M. ; Giancardo, Luca ; Chaum, Edward ; Karnowski, Thomas P. ; Mériaudeau, Fabrice. / Statistical atlas based exudate segmentation. In: Computerized Medical Imaging and Graphics. 2013 ; Vol. 37, No. 5-6. pp. 358-368.
@article{2423ac89763848bc9f43e79359eff94d,
title = "Statistical atlas based exudate segmentation",
abstract = "Diabetic macular edema (DME) is characterized by hard exudates. In this article, we propose a novel statistical atlas based method for segmentation of such exudates. Any test fundus image is first warped on the atlas co-ordinate and then a distance map is obtained with the mean atlas image. This leaves behind the candidate lesions. Post-processing schemes are introduced for final segmentation of the exudate. Experiments with the publicly available HEI-MED data-set shows good performance of the method. A lesion localization fraction of 82.5{\%} at 35{\%} of non-lesion localization fraction on the FROC curve is obtained. The method is also compared to few most recent reference methods.",
author = "Sharib Ali and D{\'e}sir{\'e} Sidib{\'e} and Adal, {Kedir M.} and Luca Giancardo and Edward Chaum and Karnowski, {Thomas P.} and Fabrice M{\'e}riaudeau",
year = "2013",
month = "7",
day = "1",
doi = "10.1016/j.compmedimag.2013.06.006",
language = "English (US)",
volume = "37",
pages = "358--368",
journal = "Computerized Medical Imaging and Graphics",
issn = "0895-6111",
publisher = "Elsevier Limited",
number = "5-6",

}

TY - JOUR

T1 - Statistical atlas based exudate segmentation

AU - Ali, Sharib

AU - Sidibé, Désiré

AU - Adal, Kedir M.

AU - Giancardo, Luca

AU - Chaum, Edward

AU - Karnowski, Thomas P.

AU - Mériaudeau, Fabrice

PY - 2013/7/1

Y1 - 2013/7/1

N2 - Diabetic macular edema (DME) is characterized by hard exudates. In this article, we propose a novel statistical atlas based method for segmentation of such exudates. Any test fundus image is first warped on the atlas co-ordinate and then a distance map is obtained with the mean atlas image. This leaves behind the candidate lesions. Post-processing schemes are introduced for final segmentation of the exudate. Experiments with the publicly available HEI-MED data-set shows good performance of the method. A lesion localization fraction of 82.5% at 35% of non-lesion localization fraction on the FROC curve is obtained. The method is also compared to few most recent reference methods.

AB - Diabetic macular edema (DME) is characterized by hard exudates. In this article, we propose a novel statistical atlas based method for segmentation of such exudates. Any test fundus image is first warped on the atlas co-ordinate and then a distance map is obtained with the mean atlas image. This leaves behind the candidate lesions. Post-processing schemes are introduced for final segmentation of the exudate. Experiments with the publicly available HEI-MED data-set shows good performance of the method. A lesion localization fraction of 82.5% at 35% of non-lesion localization fraction on the FROC curve is obtained. The method is also compared to few most recent reference methods.

UR - http://www.scopus.com/inward/record.url?scp=84887992514&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84887992514&partnerID=8YFLogxK

U2 - 10.1016/j.compmedimag.2013.06.006

DO - 10.1016/j.compmedimag.2013.06.006

M3 - Article

C2 - 23896588

AN - SCOPUS:84887992514

VL - 37

SP - 358

EP - 368

JO - Computerized Medical Imaging and Graphics

JF - Computerized Medical Imaging and Graphics

SN - 0895-6111

IS - 5-6

ER -