Towards automatic detection of abnormal retinal capillaries in ultra-widefield-of-view retinal angiographic exams

Krists Zutis; E. Trucco; J.P. Hubschman; D. Reed; S. Shah; J. Van Hemert

doi:10.1109/EMBC.2013.6611261

Towards automatic detection of abnormal retinal capillaries in ultra-widefield-of-view retinal angiographic exams

Krists Zutis, E. Trucco, J.P. Hubschman, D. Reed, S. Shah, J. Van Hemert

Research output: Chapter in Book/Report/Conference proceeding › Chapter

4 Citations (Scopus)

Abstract

Retinal capillary abnormalities include small, leaky, severely tortuous blood vessels that are associated with a variety of retinal pathologies. We present a prototype image-processing system for detecting abnormal retinal capillary regions in ultra-widefield-of-view (UWFOV) fluorescein angiography exams of the human retina. The algorithm takes as input an UWFOV FA frame and returns the candidate regions identified. An SVM classifier is trained on regions traced by expert ophthalmologists. Tests with a variety of feature sets indicate that edge features and allied properties differentiate best between normal and abnormal retinal capillary regions. Experiments with an initial set of images from patients showing branch retinal vein occlusion (BRVO) indicate promising area under the ROC curve of 0.950 and a weighted Cohen's Kappa value of 0.822.

Original language	English
Title of host publication	Proceedings of the 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS
Pages	7372-7375
Number of pages	4
DOIs	https://doi.org/10.1109/EMBC.2013.6611261
Publication status	Published - 2013

Access to Document

10.1109/EMBC.2013.6611261

Cite this

Zutis, K., Trucco, E., Hubschman, J. P., Reed, D., Shah, S., & Van Hemert, J. (2013). Towards automatic detection of abnormal retinal capillaries in ultra-widefield-of-view retinal angiographic exams. In Proceedings of the 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS (pp. 7372-7375) https://doi.org/10.1109/EMBC.2013.6611261

@inbook{de2e23b4acfd44d7a4dd0818886d2e43,

title = "Towards automatic detection of abnormal retinal capillaries in ultra-widefield-of-view retinal angiographic exams",

abstract = "Retinal capillary abnormalities include small, leaky, severely tortuous blood vessels that are associated with a variety of retinal pathologies. We present a prototype image-processing system for detecting abnormal retinal capillary regions in ultra-widefield-of-view (UWFOV) fluorescein angiography exams of the human retina. The algorithm takes as input an UWFOV FA frame and returns the candidate regions identified. An SVM classifier is trained on regions traced by expert ophthalmologists. Tests with a variety of feature sets indicate that edge features and allied properties differentiate best between normal and abnormal retinal capillary regions. Experiments with an initial set of images from patients showing branch retinal vein occlusion (BRVO) indicate promising area under the ROC curve of 0.950 and a weighted Cohen's Kappa value of 0.822.",

author = "Krists Zutis and E. Trucco and J.P. Hubschman and D. Reed and S. Shah and {Van Hemert}, J.",

year = "2013",

doi = "10.1109/EMBC.2013.6611261",

language = "English",

isbn = "9781457702167",

pages = "7372--7375",

booktitle = "Proceedings of the 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS",

}

Towards automatic detection of abnormal retinal capillaries in ultra-widefield-of-view retinal angiographic exams. / Zutis, Krists; Trucco, E.; Hubschman, J.P. et al.
Proceedings of the 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS. 2013. p. 7372-7375.

Research output: Chapter in Book/Report/Conference proceeding › Chapter

TY - CHAP

T1 - Towards automatic detection of abnormal retinal capillaries in ultra-widefield-of-view retinal angiographic exams

AU - Zutis, Krists

AU - Trucco, E.

AU - Hubschman, J.P.

AU - Reed, D.

AU - Shah, S.

AU - Van Hemert, J.

PY - 2013

Y1 - 2013

N2 - Retinal capillary abnormalities include small, leaky, severely tortuous blood vessels that are associated with a variety of retinal pathologies. We present a prototype image-processing system for detecting abnormal retinal capillary regions in ultra-widefield-of-view (UWFOV) fluorescein angiography exams of the human retina. The algorithm takes as input an UWFOV FA frame and returns the candidate regions identified. An SVM classifier is trained on regions traced by expert ophthalmologists. Tests with a variety of feature sets indicate that edge features and allied properties differentiate best between normal and abnormal retinal capillary regions. Experiments with an initial set of images from patients showing branch retinal vein occlusion (BRVO) indicate promising area under the ROC curve of 0.950 and a weighted Cohen's Kappa value of 0.822.

AB - Retinal capillary abnormalities include small, leaky, severely tortuous blood vessels that are associated with a variety of retinal pathologies. We present a prototype image-processing system for detecting abnormal retinal capillary regions in ultra-widefield-of-view (UWFOV) fluorescein angiography exams of the human retina. The algorithm takes as input an UWFOV FA frame and returns the candidate regions identified. An SVM classifier is trained on regions traced by expert ophthalmologists. Tests with a variety of feature sets indicate that edge features and allied properties differentiate best between normal and abnormal retinal capillary regions. Experiments with an initial set of images from patients showing branch retinal vein occlusion (BRVO) indicate promising area under the ROC curve of 0.950 and a weighted Cohen's Kappa value of 0.822.

UR - http://www.scopus.com/inward/record.url?scp=84886474569&partnerID=8YFLogxK

U2 - 10.1109/EMBC.2013.6611261

DO - 10.1109/EMBC.2013.6611261

M3 - Chapter

C2 - 24111448

AN - SCOPUS:84886474569

SN - 9781457702167

SP - 7372

EP - 7375

BT - Proceedings of the 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS

ER -

Towards automatic detection of abnormal retinal capillaries in ultra-widefield-of-view retinal angiographic exams

Abstract

Access to Document

Other files and links

Fingerprint

Cite this