Information Fusion for Diabetic Retinopathy CAD in Digital Color Fundus Photographs

The purpose of computer-aided detection or diagnosis (CAD) technology has so far been to serve as a second reader. If, however, all relevant lesions in an image can be detected by CAD algorithms, use of CAD for automatic reading or prescreening may become feasible. This work addresses the question how to fuse information from multiple CAD algorithms, operating on multiple images that comprise an exam, to determine a likelihood that the exam is normal and would not require further inspection by human operators. We focus on retinal image screening for diabetic retinopathy, a common complication of diabetes. Current CAD systems are not designed to automatically evaluate complete exams consisting of multiple images for which several detection algorithm output sets are available. Information fusion will potentially play a crucial role in enabling the application of CAD technology to the automatic screening problem. Several different fusion methods are proposed and their effect on the performance of a complete comprehensive automatic diabetic retinopathy screening system is evaluated. Experiments show that the choice of fusion method can have a large impact on system performance. The complete system was evaluated on a set of 15 000 exams (60 000 images). The best performing fusion method obtained an area under the receiver operator characteristic curve of 0.881. This indicates that automated prescreening could be applied in diabetic retinopathy screening programs.

[1]  G. Bresnick,et al.  A screening approach to the surveillance of patients with diabetes for the presence of vision-threatening retinopathy. , 2000, Ophthalmology.

[2]  J. Hanley,et al.  A method of comparing the areas under receiver operating characteristic curves derived from the same cases. , 1983, Radiology.

[3]  M. Abràmoff,et al.  Web-based screening for diabetic retinopathy in a primary care population: the EyeCheck project. , 2005, Telemedicine journal and e-health : the official journal of the American Telemedicine Association.

[4]  Early photocoagulation for diabetic retinopathy. ETDRS report number 9. Early Treatment Diabetic Retinopathy Study Research Group. , 1991, Ophthalmology.

[5]  D. Klonoff,et al.  An economic analysis of interventions for diabetes. , 2000, Diabetes care.

[6]  J. Boyce,et al.  Automated detection of diabetic retinopathy in digital retinal images: a tool for diabetic retinopathy screening , 2004, Diabetic medicine : a journal of the British Diabetic Association.

[7]  B. van Ginneken,et al.  Automated detection and differentiation of drusen, exudates, and cotton-wool spots in digital color fundus photographs for diabetic retinopathy diagnosis. , 2007, Investigative ophthalmology & visual science.

[8]  Chih-Jen Lin,et al.  LIBSVM: A library for support vector machines , 2011, TIST.

[9]  Peter F. Sharp,et al.  Evaluation of a System for Automatic Detection of Diabetic Retinopathy From Color Fundus Photographs in a Large Population of Patients With Diabetes , 2008, Diabetes Care.

[10]  R. Coulden,et al.  The role of preliminary interpretation of chest radiographs by radiographers in the management of acute medical problems within a cardiothoracic centre. , 2001, The British journal of radiology.

[11]  Josef Kittler,et al.  Floating search methods in feature selection , 1994, Pattern Recognit. Lett..

[12]  B. van Ginneken,et al.  Automated localization of the optic disc and the fovea , 2008, 2008 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[13]  Bram van Ginneken,et al.  Image structure clustering for image quality verification of color retina images in diabetic retinopathy screening , 2006, Medical Image Anal..

[14]  L. Aiello,et al.  Detection of diabetic macular edema. Ophthalmoscopy versus photography--Early Treatment Diabetic Retinopathy Study Report Number 5. The ETDRS Research Group. , 1989, Ophthalmology.

[15]  Bram van Ginneken,et al.  Automatic detection of red lesions in digital color fundus photographs , 2005, IEEE Transactions on Medical Imaging.

[16]  A. Renshaw,et al.  Rapid prescreening of Papanicolaou smears , 2006, Cancer.

[17]  David G. Stork,et al.  Pattern Classification (2nd ed.) , 1999 .

[18]  M. Cree,et al.  A comparison of computer based classification methods applied to the detection of microaneurysms in ophthalmic fluorescein angiograms , 1998, Comput. Biol. Medicine.

[19]  P F Sharp,et al.  Cost-effectiveness of implementing automated grading within the national screening programme for diabetic retinopathy in Scotland , 2007, British Journal of Ophthalmology.

[20]  Marco Loog,et al.  Supervised dimensionality reduction and contextual pattern recognition in medical image processing , 2004 .

[21]  M.D. Abramoff,et al.  The automatic detection of the optic disc location in retinal images using optic disc location regression , 2006, 2006 International Conference of the IEEE Engineering in Medicine and Biology Society.

[22]  Peter F Sharp,et al.  A fully automated comparative microaneurysm digital detection system , 1997, Eye.

[23]  P F Sharp,et al.  An image-processing strategy for the segmentation and quantification of microaneurysms in fluorescein angiograms of the ocular fundus. , 1996, Computers and biomedical research, an international journal.

[24]  Sunil Arya,et al.  An optimal algorithm for approximate nearest neighbor searching fixed dimensions , 1998, JACM.

[25]  Max A. Viergever,et al.  Ridge-based vessel segmentation in color images of the retina , 2004, IEEE Transactions on Medical Imaging.

[26]  J. Olson,et al.  The efficacy of automated “disease/no disease” grading for diabetic retinopathy in a systematic screening programme , 2007, British Journal of Ophthalmology.

[27]  Bram van Ginneken,et al.  Comparative study of retinal vessel segmentation methods on a new publicly available database , 2004, SPIE Medical Imaging.

[28]  Nico Karssemeijer,et al.  Computer-Aided Diagnosis in Medical Imaging , 2001, IEEE Trans. Medical Imaging.