Journal article
Image structure clustering for image quality verification of color retina images in diabetic retinopathy screening
Medical Image Analysis, Vol.10(6), pp.888-898
2006
DOI: 10.1016/j.media.2006.09.006
PMID: 17138215
Abstract
Reliable verification of image quality of retinal screening images is a prerequisite for the development of automatic screening systems for diabetic retinopathy. A system is presented that can automatically determine whether the quality of a retinal screening image is sufficient for automatic analysis. The system is based on the assumption that an image of sufficient quality should contain particular image structures according to a certain pre-defined distribution. We cluster filterbank response vectors to obtain a compact representation of the image structures found within an image. Using this compact representation together with raw histograms of the R, G, and B color planes, a statistical classifier is trained to distinguish normal from low quality images. The presented system does not require any previous segmentation of the image in contrast with previous work. The system was evaluated on a large, representative set of 1000 images obtained in a screening program. The proposed method, using different feature sets and classifiers, was compared with the ratings of a second human observer. The best system, based on a Support Vector Machine, has performance close to optimal with an area under the ROC curve of 0.9968.
Details
- Title: Subtitle
- Image structure clustering for image quality verification of color retina images in diabetic retinopathy screening
- Creators
- Meindert Niemeijer - University Medical Center Utrecht, Image Sciences Institute, Q0S.459, Heidelberglaan 100, 3584CX, Utrecht, The NetherlandsMichael D Abràmoff - Department of Ophthalmology and Visual Sciences, University of Iowa Hospitals and Clinics, 200 Hawkins Drive, Iowa City, IA 52242, USABram van Ginneken - University Medical Center Utrecht, Image Sciences Institute, Q0S.459, Heidelberglaan 100, 3584CX, Utrecht, The Netherlands
- Resource Type
- Journal article
- Publication Details
- Medical Image Analysis, Vol.10(6), pp.888-898
- Publisher
- Elsevier B.V
- DOI
- 10.1016/j.media.2006.09.006
- PMID
- 17138215
- ISSN
- 1361-8415
- eISSN
- 1361-8423
- Language
- English
- Date published
- 2006
- Academic Unit
- Roy J. Carver Department of Biomedical Engineering; Electrical and Computer Engineering; Ophthalmology and Visual Sciences
- Record Identifier
- 9983806280102771
Metrics
29 Record Views