Dataset Information

Detection of diabetic retinopathy using a fusion of textural and ridgelet features of retinal images and sequential minimal optimization classifier.

ABSTRACT: Diabetes is one of the most prevalent diseases in the world, which is a metabolic disorder characterized by high blood sugar. Diabetes complications are leading to Diabetic Retinopathy (DR). The early stages of DR may have either no sign or cause minor vision problems, but later stages of the disease can lead to blindness. DR diagnosis is an exceedingly difficult task because of changes in the retina during the disease stages. An automatic DR early detection method can save a patient's vision and can also support the ophthalmologists in DR screening. This paper develops a model for the diagnostics of DR. Initially, we extract and fuse the ophthalmoscopic features from the retina images based on textural gray-level features like co-occurrence, run-length matrix, as well as the coefficients of the Ridgelet Transform. Based on the retina features, the Sequential Minimal Optimization (SMO) classification is used to classify diabetic retinopathy. For performance analysis, the openly accessible retinal image datasets are used, and the findings of the experiments demonstrate the quality and efficacy of the proposed method (we achieved 98.87% sensitivity, 95.24% specificity, 97.05% accuracy on DIARETDB1 dataset, and 90.9% sensitivity, 91.0% specificity, 91.0% accuracy on KAGGLE dataset).

SUBMITTER: Ramasamy LK

PROVIDER: S-EPMC8114804 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Importance:Quantitative measurements based on optical coherence tomographic angiography (OCTA) may have value in managing diabetic retinopathy (DR), but there is limited information on the ability of OCTA to distinguish eyes with DR. Objective:To evaluate the ability of measurements of retinal microvasculature using OCTA to distinguish healthy eyes from eyes with DR. Design, Setting, and Participants:In this prospective cross-sectional study, OCTA was used to examine the eyes of participants with type 2 diabetes with or without DR and the eyes of participants without diabetes from September 17, 2015, to April 6, 2016. Density maps based on superficial retinal layer (SRL) and deeper retinal layer (DRL) images were generated after a method to remove decorrelation tails was applied to the DRL images. Exposures:Both eyes of each participant were examined by means of a 3-mm OCTA scan and 7-field fundus photography using the Diabetic Retinopathy Severity Scale. Main Outcomes and Measures:Two measures were examined: perfusion density, based on the area of vessels, and vessel density, based on a map with vessels of 1-pixel width. The size of the foveal avascular zone was also calculated automatically, and so was the area under the receiver operating characteristic curve. Results:A total of 50 eyes from 26 participants with diabetes (10 women and 16 men; mean [SD] age, 64.9 [7.5] years) and 50 healthy eyes from 25 participants without diabetes (14 women and 11 men; mean [SD] age, 64.0 [7.1] years) were imaged. All participants were white. Vessel density measured in the SRL had the highest area under the receiver operating characteristic curve (0.893 [95% CI, 0.827-0.959]), compared with perfusion density in the SRL (0.794 [95% CI, 0.707-0.881]), foveal avascular zone area (0.472 [95% CI, 0.356-0.588]), and vessel density in the DRL (0.703 [95% CI, 0.601-0.805]). Vessel density in the SRL negatively correlated with best-corrected visual acuity (r?=?-0.28; P?=?.05) and severity of DR (r?=?-0.46; P?=?.001). Density metrics correlated with age. No correlation was detected between vascular density or foveal avascular zone metrics and hemoglobin A1C or duration of diabetes. Conclusions and Relevance:Vessel density measured by OCTA provides a quantitative metric of capillary closure that correlates with severity of DR and may allow staging, diagnosis, and monitoring that do not require subjective evaluation of fundus images.

Project description:PurposeTo evaluate the diagnostic accuracy of machine learning (ML) techniques applied to radiomic features extracted from OCT and OCT angiography (OCTA) images for diabetes mellitus (DM), diabetic retinopathy (DR), and referable DR (R-DR) diagnosis.DesignCross-sectional analysis of a retinal image dataset from a previous prospective OCTA study (ClinicalTrials.govNCT03422965).ParticipantsPatients with type 1 DM and controls included in the progenitor study.MethodsRadiomic features were extracted from fundus retinographies, OCT, and OCTA images in each study eye. Logistic regression, linear discriminant analysis, support vector classifier (SVC)-linear, SVC-radial basis function, and random forest models were created to evaluate their diagnostic accuracy for DM, DR, and R-DR diagnosis in all image types.Main outcome measuresArea under the receiver operating characteristic curve (AUC) mean and standard deviation for each ML model and each individual and combined image types.ResultsA dataset of 726 eyes (439 individuals) were included. For DM diagnosis, the greatest AUC was observed for OCT (0.82, 0.03). For DR detection, the greatest AUC was observed for OCTA (0.77, 0.03), especially in the 3 × 3 mm superficial capillary plexus OCTA scan (0.76, 0.04). For R-DR diagnosis, the greatest AUC was observed for OCTA (0.87, 0.12) and the deep capillary plexus OCTA scan (0.86, 0.08). The addition of clinical variables (age, sex, etc.) improved most models AUC for DM, DR and R-DR diagnosis. The performance of the models was similar in unilateral and bilateral eyes image datasets.ConclusionsRadiomics extracted from OCT and OCTA images allow identification of patients with DM, DR, and R-DR using standard ML classifiers. OCT was the best test for DM diagnosis, OCTA for DR and R-DR diagnosis and the addition of clinical variables improved most models. This pioneer study demonstrates that radiomics-based ML techniques applied to OCT and OCTA images may be an option for DR screening in patients with type 1 DM.Financial disclosuresProprietary or commercial disclosure may be found after the references.

Dataset Information

Detection of diabetic retinopathy using a fusion of textural and ridgelet features of retinal images and sequential minimal optimization classifier.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets