DETECTION AND LOCALIZATION OF RETINAL EXUDATES FOR DIABETIC RETINOPATHY

2015 ◽  
Vol 23 (02) ◽  
pp. 195-212 ◽  
Author(s):  
G. MAHENDRAN ◽  
R. DHANASEKARAN
Keyword(s):  
Diabetic Retinopathy ◽  
Sensitivity And Specificity ◽  
Retinal Images ◽  
Chemical Solution ◽  
Automated Method ◽  
Retinal Capillary ◽  
Fcm Clustering ◽  
Image Processing Algorithms ◽  
Detection And Localization ◽  
Retinal Exudates

Diabetic retinopathy (DR) is a complication of diabetes caused by changes in the blood vessels of the retina. Initially, the DR causes trivial changes in the retinal capillary. The symptoms can blur or distort patients' vision, which are the main causes of blindness. The DR is characterized by the presence of exudates at the nonproliferative stage. Once damaged by DR, the effects will be permanent and hence an earlier treatment is considered as vital. The presence of exudates is detected by ophthalmologists from the dilated retinal images, which are captured by dropping chemical solution into the patient's eye that leads to irritation. Therefore, there is a need for an alternative method toward the detection of exudates using image processing algorithms from the nondilated images. In this paper, an automated method is proposed for the detection of exudates using the fuzzy C-Means (FCM) clustering technique and reconstruction through a superimposition process in the absence of dilating patient's eye. The segmented result of FCM is compared with the result obtained using the Fuzzy K-Means segmentation algorithm. The sensitivity and specificity values for the exudates detection using the FCM algorithm are 87.38% and 96.94%, respectively. On the other hand, sensitivity and specificity values for the exudates detection using the K-Means algorithm are 75.04% and 93.73%, respectively.

scholarly journals Automated Detection of Diabetic Retinopathy for Early Diagnosis using Exudate Images

2019 ◽  
Vol 8 (6S3) ◽  
pp. 1572-1576
Keyword(s):  
Diabetic Retinopathy ◽  
Early Diagnosis ◽  
Sensitivity And Specificity ◽  
Automatic Recognition ◽  
Automated Detection ◽  
Retinal Images ◽  
Feature Classification ◽  
Detection Process ◽  
Performance Of Algorithm ◽  
Retinal Exudates

Retina plays a vital character in detection of various diseases in early point such as diabetes retinopathy which can be performed by analyzing the retinal images [6]. Diseased patients have to undergo periodic screening of eye. Standouts amongst the most predominant clinical indications of diabetic retinopathy are exudates [17]. To detect diabetic retinopathy in patients the ophthalmologist inspects the exudates by Ophthalmoscopy [17] where recognition of exudates is a vital diagnostic undertaking in which computer help may assume a noteworthy job. But intrinsic characteristics of retinal images detection process is difficult for the ophthalmologists. Here, we proposed another algorithm “Superpixel Multi-Feature Classification" for the programmed automatic recognition of retinal exudates successfully and to encourage ophthalmologist to give better patient finding experiencing diabetic retinopathy, advising them the level of seriousness ahead of time. The performance of algorithm has been compared as a result, the outcomes are effective and the sensitivity and specificity for our exudates identification is 80% and 91.28%, respectively [15].

scholarly journals Simple, Mobile-based Artificial Intelligence Algorithm in the detection of Diabetic Retinopathy (SMART) study

2020 ◽  
Vol 8 (1) ◽  
pp. e000892 ◽  
Author(s):  
Bhavana Sosale ◽  
Sosale Ramachandra Aravind ◽  
Hemanth Murthy ◽  
Srikanth Narayana ◽  
Usha Sharma ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Diabetic Retinopathy ◽  
Sensitivity And Specificity ◽  
Smart Phone ◽  
High Sensitivity ◽  
Ground Truth ◽  
Cross Sectional Study ◽  
Retinal Images ◽  
Cross Sectional ◽  
Smart Study

IntroductionThe aim of this study is to evaluate the performance of the offline smart phone-based Medios artificial intelligence (AI) algorithm in the diagnosis of diabetic retinopathy (DR) using non-mydriatic (NM) retinal images.MethodsThis cross-sectional study prospectively enrolled 922 individuals with diabetes mellitus. NM retinal images (disc and macula centered) from each eye were captured using the Remidio NM fundus-on-phone (FOP) camera. The images were run offline and the diagnosis of the AI was recorded (DR present or absent). The diagnosis of the AI was compared with the image diagnosis of five retina specialists (majority diagnosis considered as ground truth).ResultsAnalysis included images from 900 individuals (252 had DR). For any DR, the sensitivity and specificity of the AI algorithm was found to be 83.3% (95% CI 80.9% to 85.7%) and 95.5% (95% CI 94.1% to 96.8%). The sensitivity and specificity of the AI algorithm in detecting referable DR (RDR) was 93% (95% CI 91.3% to 94.7%) and 92.5% (95% CI 90.8% to 94.2%).ConclusionThe Medios AI has a high sensitivity and specificity in the detection of RDR using NM retinal images.

scholarly journals Detection of Subclinical Diabetic Retinopathy by Fine Structure Analysis of Retinal Images

2019 ◽  
Vol 2019 ◽  
pp. 1-6 ◽  
Author(s):  
Maziyar M. Khansari ◽  
William D. O’Neill ◽  
Richard D. Penn ◽  
Norman P. Blair ◽  
Mahnaz Shahidi
Keyword(s):  
Diabetic Retinopathy ◽  
Fine Structure ◽  
Structure Analysis ◽  
Ordinary Least Squares ◽  
Retinal Images ◽  
Human Observer ◽  
Linear Discriminant ◽  
Automated Method ◽  
Fine Structure Analysis ◽  
Discrimination Rate

Background and Objective. Diabetic retinopathy (DR) is a major complication of diabetes and the leading cause of blindness among US working-age adults. Detection of subclinical DR is important for disease monitoring and prevention of damage to the retina before occurrence of vision loss. The purpose of this retrospective study is to describe an automated method for discrimination of subclinical DR using fine structure analysis of retinal images. Methods. Discrimination between nondiabetic control (NC; N = 16) and diabetic without clinical retinopathy (NDR; N = 17) subjects was performed using ordinary least squares regression and Fisher’s linear discriminant analysis. A human observer also performed the discrimination by visual inspection of the images. Results. The discrimination rate for subclinical DR was 88% using the automated method and higher than the rate obtained by a human observer which was 45%. Conclusions. The method provides sensitive and rapid analysis of retinal images and could be useful in detecting subclinical DR.

Sensitivity and specificity of digital retinal images in grading diabetic retinopathy

2004 ◽  
Vol 82 (2) ◽  
pp. 126-130 ◽  
Author(s):  
Jukka M. Saari ◽  
Paula Summanen ◽  
Tero Kivelä ◽  
K. Matti Saari
Keyword(s):  
Diabetic Retinopathy ◽  
Sensitivity And Specificity ◽  
Retinal Images

scholarly journals An Automatic Localization of Optic Disc in Low Resolution Retinal Images by Modified Directional Matched Filter

2018 ◽  
Vol 16 (1) ◽  
pp. 1-7
Author(s):  
Murugan Raman ◽  
Reeba Korah ◽  
Kavitha Tamilselvan
Keyword(s):  
Diabetic Retinopathy ◽  
Optic Disc ◽  
Matched Filter ◽  
Retinal Images ◽  
Low Resolution ◽  
Automated Method ◽  
Retinal Blood Vessels ◽  
Optic Disc Localization ◽  
Automatic Localization ◽  
Success Percentage

An automatic optic disc localization in retinal images used to screen eye related diseases like diabetic retinopathy. Many techniques are available to detect Optic Disc (OD) in high-resolution retinal images. Unfortunately, there are no efficient methods available to detect OD in low-resolution retinal images. The objective of this research paper is to develop an automated method for localization of Optic Disc in low resolution retinal images. This paper proposes a modified directional matched filter parameters of the retinal blood vessels to localize the center of optic disc. The proposed method was implemented in MATLAB and evaluated both normal and abnormal low resolution retinal images using the subset of Optic Nerve Head Segmentation Dataset (ONHSD) and the success percentage was found to be an average of 96.96% with 23seconds

scholarly journals Exudates as Landmarks Identified through FCM Clustering in Retinal Images

2020 ◽  
Vol 11 (1) ◽  
pp. 142
Author(s):  
Hadi Hamad ◽  
Tahreer Dwickat ◽  
Domenico Tegolo ◽  
Cesare Valenti
Keyword(s):  
Diabetic Retinopathy ◽  
Random Selection ◽  
Retinal Images ◽  
Automatic Identification ◽  
Fundus Images ◽  
Fuzzy C Means Clustering ◽  
Fcm Clustering ◽  
Clustering Approach ◽  
Sensitivity Specificity ◽  
Selection Of

The aim of this work was to develop a method for the automatic identification of exudates, using an unsupervised clustering approach. The ability to classify each pixel as belonging to an eventual exudate, as a warning of disease, allows for the tracking of a patient’s status through a noninvasive approach. In the field of diabetic retinopathy detection, we considered four public domain datasets (DIARETDB0/1, IDRID, and e-optha) as benchmarks. In order to refine the final results, a specialist ophthalmologist manually segmented a random selection of DIARETDB0/1 fundus images that presented exudates. An innovative pipeline of morphological procedures and fuzzy C-means clustering was integrated in order to extract exudates with a pixel-wise approach. Our methodology was optimized, and verified and the parameters were fine-tuned in order to define both suitable values and to produce a more accurate segmentation. The method was used on 100 tested images, resulting in averages of sensitivity, specificity, and accuracy equal to 83.3%, 99.2%, and 99.1%, respectively.

Sign in / Sign up

Export Citation Format

Share Document