Automated image quality evaluation of retinal fundus photographs in diabetic retinopathy screening

This paper presents a system that can automatically determine whether the quality of a retinal image is sufficient for computer-based diabetic retinopathy (DR) screening. The system integrates global histogram features, textural features, and vessel density, as well as a local non-reference perceptual sharpness metric. A partial least square (PLS) classifier is trained to distinguish low quality images from normal quality images. The system was evaluated on a large, representative set of 1884 non-mydriatic retinal images from 412 subjects. An area under the ROC curve of 96% was achieved.

[1]  Langis Gagnon,et al.  Automatic visual quality assessment in optical fundus images , 2001 .

[2]  Joachim Hornegger,et al.  Automated quality assessment of retinal fundus photos , 2010, International Journal of Computer Assisted Radiology and Surgery.

[3]  M. Obermaier,et al.  A telemedical approach to the screening of diabetic retinopathy: digital fundus photography. , 2000, Diabetes care.

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

[5]  Lina J. Karam,et al.  A No-Reference Image Blur Metric Based on the Cumulative Probability of Blur Detection (CPBD) , 2011, IEEE Transactions on Image Processing.

[6]  J. Olson,et al.  Automated assessment of diabetic retinal image quality based on clarity and field definition. , 2006, Investigative ophthalmology & visual science.

[7]  T. Teng,et al.  Progress towards automated diabetic ocular screening: A review of image analysis and intelligent systems for diabetic retinopathy , 2006, Medical and Biological Engineering and Computing.

[8]  Marios S. Pattichis,et al.  Multiscale AM-FM Methods for Diabetic Retinopathy Lesion Detection , 2010, IEEE Transactions on Medical Imaging.

[9]  E. Simon Barriga,et al.  Vision-based, real-time retinal image quality assessment , 2009, 2009 22nd IEEE International Symposium on Computer-Based Medical Systems.

[10]  Lina J. Karam,et al.  A No-Reference Objective Image Sharpness Metric Based on the Notion of Just Noticeable Blur (JNB) , 2009, IEEE Transactions on Image Processing.

[11]  Yiming Wang,et al.  Automatic retinal image quality assessment and enhancement , 1999, Medical Imaging.

[12]  Gilberto Zamora,et al.  Fast vessel segmentation in retinal images using multi-scale enhancement and second-order local entropy , 2012, Medical Imaging.

[13]  J. Olson,et al.  Automated grading for diabetic retinopathy: a large-scale audit using arbitration by clinical experts , 2010, British Journal of Ophthalmology.

[14]  Mohamed-Chaker Larabi,et al.  No-reference perceptual blur metric for stereoscopic images , 2014, 2014 International Conference on 3D Imaging (IC3D).

[15]  P. Scanlon,et al.  Comparison of two reference standards in validating two field mydriatic digital photography as a method of screening for diabetic retinopathy , 2003, The British journal of ophthalmology.

[16]  K. W. Tobin,et al.  Elliptical local vessel density: A fast and robust quality metric for retinal images , 2008, 2008 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[17]  J A Olson,et al.  The impact of the Health Technology Board for Scotland’s grading model on referrals to ophthalmology services , 2005, British Journal of Ophthalmology.

[18]  Robert M. Haralick,et al.  Textural Features for Image Classification , 1973, IEEE Trans. Syst. Man Cybern..