Automated classifiers for early detection and diagnosis of retinopathy in diabetic eyes

BackgroundArtificial neural networks (ANNs) have been used to classify eye diseases, such as diabetic retinopathy (DR) and glaucoma. DR is the leading cause of blindness in working-age adults in the developed world. The implementation of DR diagnostic routines could be feasibly improved by the integration of structural and optical property test measurements of the retinal structure that provide important and complementary information for reaching a diagnosis. In this study, we evaluate the capability of several structural and optical features (thickness, total reflectance and fractal dimension) of various intraretinal layers extracted from optical coherence tomography images to train a Bayesian ANN to discriminate between healthy and diabetic eyes with and with no mild retinopathy.ResultsWhen exploring the probability as to whether the subject’s eye was healthy (diagnostic condition, Test 1), we found that the structural and optical property features of the outer plexiform layer (OPL) and the complex formed by the ganglion cell and inner plexiform layers (GCL + IPL) provided the highest probability (positive predictive value (PPV) of 91% and 89%, respectively) for the proportion of patients with positive test results (healthy condition) who were correctly diagnosed (Test 1). The true negative, TP and PPV values remained stable despite the different sizes of training data sets (Test 2). The sensitivity, specificity and PPV were greater or close to 0.70 for the retinal nerve fiber layer’s features, photoreceptor outer segments and retinal pigment epithelium when 23 diabetic eyes with mild retinopathy were mixed with 38 diabetic eyes with no retinopathy (Test 3).ConclusionsA Bayesian ANN trained on structural and optical features from optical coherence tomography data can successfully discriminate between healthy and diabetic eyes with and with no retinopathy. The fractal dimension of the OPL and the GCL + IPL complex predicted by the Bayesian radial basis function network provides better diagnostic utility to classify diabetic eyes with mild retinopathy. Moreover, the thickness and fractal dimension parameters of the retinal nerve fiber layer, photoreceptor outer segments and retinal pigment epithelium show promise for the diagnostic classification between diabetic eyes with and with no mild retinopathy.

[1]  M F Marmor,et al.  The interphotoreceptor matrix mediates primate retinal adhesion. , 1995, Archives of ophthalmology.

[2]  Delia Cabrera DeBuc,et al.  Reliability and reproducibility of macular segmentation using a custom-built optical coherence tomography retinal image analysis software. , 2009, Journal of biomedical optics.

[3]  Mark A. Musen,et al.  Medical applications of artificial neural networks : connectionist models of survival , 1996 .

[4]  K Nakamura,et al.  Effect of an artificial neural network on radiologists' performance in the differential diagnosis of interstitial lung disease using chest radiographs. , 1999, AJR. American journal of roentgenology.

[5]  Wei Gao,et al.  Improving image segmentation performance and quantitative analysis via a computer-aided grading methodology for optical coherence tomography retinal image analysis. , 2010, Journal of biomedical optics.

[6]  U. Rajendra Acharya,et al.  Identification of different stages of diabetic retinopathy using retinal optical images , 2008, Inf. Sci..

[7]  Carmen A Puliafito,et al.  Automated detection of retinal layer structures on optical coherence tomography images. , 2005, Optics express.

[8]  J. Shaw,et al.  Global estimates of the prevalence of diabetes for 2010 and 2030. , 2010, Diabetes research and clinical practice.

[9]  Wei Gao,et al.  Assessing the Performance of Optical Properties Determination of Intraretinal Layers in Healthy Normal and Type 1 Diabetic Eyes using Optical Coherence Tomography , 2011 .

[10]  D L DeMets,et al.  The Wisconsin epidemiologic study of diabetic retinopathy. III. Prevalence and risk of diabetic retinopathy when age at diagnosis is 30 or more years. , 1984, Archives of ophthalmology.

[11]  Yuichi Nishiyama,et al.  Development of novel mucoadhesive pellets of metformin hydrochloride , 2009, Archives of pharmacal research.

[12]  T. Williamson,et al.  Automatic detection of diabetic retinopathy using an artificial neural network: a screening tool. , 1996, The British journal of ophthalmology.

[13]  James G. Fujimoto,et al.  Stratus OCT Image Quality Assessment , 2004 .

[14]  Chanjira Sinthanayothin,et al.  Automated screening system for diabetic retinopathy , 2003, 3rd International Symposium on Image and Signal Processing and Analysis, 2003. ISPA 2003. Proceedings of the.

[15]  Sobha Sivaprasad,et al.  Retinal Neuronal Changes In People With Diabetes , 2012 .

[16]  Richard S. Sutton,et al.  Neural networks for control , 1990 .

[17]  D L DeMets,et al.  The Wisconsin Epidemiologic Study of Diabetic Retinopathy. VI. Retinal photocoagulation. , 1987, Ophthalmology.

[18]  J W Moul,et al.  Applications of neural networks in urologic oncology. , 1998, Seminars in urologic oncology.

[19]  Haogang Zhu,et al.  Quantifying discordance between structure and function measurements in the clinical assessment of glaucoma. , 2011, Archives of ophthalmology.

[20]  Yoshinori Mitamura,et al.  Optical coherence tomography for complete management of patients with diabetic retinopathy. , 2010, Current diabetes reviews.

[21]  BMC Bioinformatics , 2005 .

[22]  Haogang Zhu,et al.  Predicting visual function from the measurements of retinal nerve fiber layer structure. , 2010, Investigative ophthalmology & visual science.

[23]  Te-Won Lee,et al.  Bayesian machine learning classifiers for combining structural and functional measurements to classify healthy and glaucomatous eyes. , 2008, Investigative ophthalmology & visual science.

[24]  Maciej Wojtkowski,et al.  Retinal assessment using optical coherence tomography , 2006, Progress in Retinal and Eye Research.

[25]  Davide Dazzi,et al.  Classification and prediction of the progression of thyroid-associated ophthalmopathy by an artificial neural network. , 2002, Ophthalmology.

[26]  Gábor Márk Somfai,et al.  Early detection of retinal thickness changes in diabetes using Optical Coherence Tomography. , 2010, Medical science monitor : international medical journal of experimental and clinical research.

[27]  A E Leure-duPree,et al.  Electron-opaque inclusions in the rat retinal pigment epithelium after treatment with chelators of zinc. , 1981, Investigative ophthalmology & visual science.

[28]  Ricardo S Silva.  Source , 2000, BMJ : British Medical Journal.

[29]  S. Agatonovic-Kustrin,et al.  Basic concepts of artificial neural network (ANN) modeling and its application in pharmaceutical research. , 2000, Journal of pharmaceutical and biomedical analysis.

[30]  Sobha Sivaprasad,et al.  Prevalence of diabetic retinopathy in various ethnic groups: a worldwide perspective. , 2012, Survey of ophthalmology.

[31]  P. Lapuerta,et al.  Neural Network Assessment of Perioperative Cardiac Risk in Vascular Surgery Patients , 1998, Medical decision making : an international journal of the Society for Medical Decision Making.

[32]  Heekuck Oh,et al.  Neural Networks for Pattern Recognition , 1993, Adv. Comput..

[33]  R. Raman,et al.  Is neuronal dysfunction an early sign of diabetic retinopathy? Microperimetry and Spectral Domain Optical Coherence Tomography (SD-OCT) Study in individuals with diabetes, but no diabetic retinopathy , 2009, Eye.