Automatic detection of retina disease: Robustness to image quality and localization of anatomy structure

T. P. Karnowski, D. Aykac, L. Giancardo, Y. Li, T. Nichols, K. W. Tobin, Edward Chaum

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The automated detection of diabetic retinopathy and other eye diseases in images of the retina has great promise as a low-cost method for broad-based screening. Many systems in the literature which perform automated detection include a quality estimation step and physiological feature detection, including the vascular tree and the optic nerve / macula location. In this work, we study the robustness of an automated disease detection method with respect to the accuracy of the optic nerve location and the quality of the images obtained as judged by a quality estimation algorithm. The detection algorithm features microaneurysm and exudate detection followed by feature extraction on the detected population to describe the overall retina image. Labeled images of retinas ground-truthed to disease states are used to train a supervised learning algorithm to identify the disease state of the retina image and exam set. Under the restrictions of high confidence optic nerve detections and good quality imagery, the system achieves a sensitivity and specificity of 94.8% and 78.7% with area-under-curve of 95.3%. Analysis of the effect of constraining quality and the distinction between mild non-proliferative diabetic retinopathy, normal retina images, and more severe disease states is included.

Original languageEnglish (US)
Title of host publication33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2011
Pages5959-5964
Number of pages6
DOIs
StatePublished - Dec 26 2011
Event33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2011 - Boston, MA, United States
Duration: Aug 30 2011Sep 3 2011

Publication series

NameProceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS
ISSN (Print)1557-170X

Other

Other33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2011
CountryUnited States
CityBoston, MA
Period8/30/119/3/11

Fingerprint

Image quality
Retina
Anatomy
Optic Nerve
Optics
Diabetic Retinopathy
Eye Diseases
Imagery (Psychotherapy)
Supervised learning
Exudates and Transudates
Learning algorithms
Area Under Curve
Blood Vessels
Feature extraction
Screening
Learning
Costs and Cost Analysis
Sensitivity and Specificity
Population
Costs

All Science Journal Classification (ASJC) codes

  • Signal Processing
  • Biomedical Engineering
  • Computer Vision and Pattern Recognition
  • Health Informatics

Cite this

Karnowski, T. P., Aykac, D., Giancardo, L., Li, Y., Nichols, T., Tobin, K. W., & Chaum, E. (2011). Automatic detection of retina disease: Robustness to image quality and localization of anatomy structure. In 33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2011 (pp. 5959-5964). [6091473] (Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS). https://doi.org/10.1109/IEMBS.2011.6091473

Automatic detection of retina disease : Robustness to image quality and localization of anatomy structure. / Karnowski, T. P.; Aykac, D.; Giancardo, L.; Li, Y.; Nichols, T.; Tobin, K. W.; Chaum, Edward.

33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2011. 2011. p. 5959-5964 6091473 (Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Karnowski, TP, Aykac, D, Giancardo, L, Li, Y, Nichols, T, Tobin, KW & Chaum, E 2011, Automatic detection of retina disease: Robustness to image quality and localization of anatomy structure. in 33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2011., 6091473, Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS, pp. 5959-5964, 33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2011, Boston, MA, United States, 8/30/11. https://doi.org/10.1109/IEMBS.2011.6091473
Karnowski TP, Aykac D, Giancardo L, Li Y, Nichols T, Tobin KW et al. Automatic detection of retina disease: Robustness to image quality and localization of anatomy structure. In 33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2011. 2011. p. 5959-5964. 6091473. (Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS). https://doi.org/10.1109/IEMBS.2011.6091473
Karnowski, T. P. ; Aykac, D. ; Giancardo, L. ; Li, Y. ; Nichols, T. ; Tobin, K. W. ; Chaum, Edward. / Automatic detection of retina disease : Robustness to image quality and localization of anatomy structure. 33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2011. 2011. pp. 5959-5964 (Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS).
@inproceedings{7f15e76d7a684384b7e58719da2364bc,
title = "Automatic detection of retina disease: Robustness to image quality and localization of anatomy structure",
abstract = "The automated detection of diabetic retinopathy and other eye diseases in images of the retina has great promise as a low-cost method for broad-based screening. Many systems in the literature which perform automated detection include a quality estimation step and physiological feature detection, including the vascular tree and the optic nerve / macula location. In this work, we study the robustness of an automated disease detection method with respect to the accuracy of the optic nerve location and the quality of the images obtained as judged by a quality estimation algorithm. The detection algorithm features microaneurysm and exudate detection followed by feature extraction on the detected population to describe the overall retina image. Labeled images of retinas ground-truthed to disease states are used to train a supervised learning algorithm to identify the disease state of the retina image and exam set. Under the restrictions of high confidence optic nerve detections and good quality imagery, the system achieves a sensitivity and specificity of 94.8{\%} and 78.7{\%} with area-under-curve of 95.3{\%}. Analysis of the effect of constraining quality and the distinction between mild non-proliferative diabetic retinopathy, normal retina images, and more severe disease states is included.",
author = "Karnowski, {T. P.} and D. Aykac and L. Giancardo and Y. Li and T. Nichols and Tobin, {K. W.} and Edward Chaum",
year = "2011",
month = "12",
day = "26",
doi = "10.1109/IEMBS.2011.6091473",
language = "English (US)",
isbn = "9781424441211",
series = "Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS",
pages = "5959--5964",
booktitle = "33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2011",

}

TY - GEN

T1 - Automatic detection of retina disease

T2 - Robustness to image quality and localization of anatomy structure

AU - Karnowski, T. P.

AU - Aykac, D.

AU - Giancardo, L.

AU - Li, Y.

AU - Nichols, T.

AU - Tobin, K. W.

AU - Chaum, Edward

PY - 2011/12/26

Y1 - 2011/12/26

N2 - The automated detection of diabetic retinopathy and other eye diseases in images of the retina has great promise as a low-cost method for broad-based screening. Many systems in the literature which perform automated detection include a quality estimation step and physiological feature detection, including the vascular tree and the optic nerve / macula location. In this work, we study the robustness of an automated disease detection method with respect to the accuracy of the optic nerve location and the quality of the images obtained as judged by a quality estimation algorithm. The detection algorithm features microaneurysm and exudate detection followed by feature extraction on the detected population to describe the overall retina image. Labeled images of retinas ground-truthed to disease states are used to train a supervised learning algorithm to identify the disease state of the retina image and exam set. Under the restrictions of high confidence optic nerve detections and good quality imagery, the system achieves a sensitivity and specificity of 94.8% and 78.7% with area-under-curve of 95.3%. Analysis of the effect of constraining quality and the distinction between mild non-proliferative diabetic retinopathy, normal retina images, and more severe disease states is included.

AB - The automated detection of diabetic retinopathy and other eye diseases in images of the retina has great promise as a low-cost method for broad-based screening. Many systems in the literature which perform automated detection include a quality estimation step and physiological feature detection, including the vascular tree and the optic nerve / macula location. In this work, we study the robustness of an automated disease detection method with respect to the accuracy of the optic nerve location and the quality of the images obtained as judged by a quality estimation algorithm. The detection algorithm features microaneurysm and exudate detection followed by feature extraction on the detected population to describe the overall retina image. Labeled images of retinas ground-truthed to disease states are used to train a supervised learning algorithm to identify the disease state of the retina image and exam set. Under the restrictions of high confidence optic nerve detections and good quality imagery, the system achieves a sensitivity and specificity of 94.8% and 78.7% with area-under-curve of 95.3%. Analysis of the effect of constraining quality and the distinction between mild non-proliferative diabetic retinopathy, normal retina images, and more severe disease states is included.

UR - http://www.scopus.com/inward/record.url?scp=84862288019&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84862288019&partnerID=8YFLogxK

U2 - 10.1109/IEMBS.2011.6091473

DO - 10.1109/IEMBS.2011.6091473

M3 - Conference contribution

C2 - 22255697

AN - SCOPUS:84862288019

SN - 9781424441211

T3 - Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS

SP - 5959

EP - 5964

BT - 33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2011

ER -