A recursive support vector machine (RSVM) algorithm to detect and classify diabetic retinopathy in fundus retina images

If the retinal images show evidences of abnormalities such as change in volume, diameter, and unusual spots in the retina, then there is a positive correlation to the diabetic progress. Mathematical and statistical theories behind the machine learning algorithms are powerful enough to detect signs of diabetes through retinal images. Several machine learning algorithms: Logistic Regression, Support Vector Machine, Random Forest, and Neural Networks were applied to predict whether images contain signs of diabetic retinopathy or not. After building the models, the computed results of these algorithms were compared by confusion matrixes, receiver operating characteristic curves, and Precision-Recall curves. The performance of the Support Vector Machine algorithm was the best since it had the highest true-positive rate, area under the curve for ROC curve, and area under the curve for Precision-Recall curve. This conclusion shows that the most complex algorithms doesn’t always give the best performance, the final accuracy also depends on the dataset. For this dataset of retinal imaging, the Support Vector Machine algorithm achieved the best results. Detecting signs of diabetic retinopathy is helpful for detecting for diabetes since more than 60% of patients with diabetes have signs of diabetic retinopathy. Machine learning algorithms can speed up the process and improve the accuracy of diagnosis. When the method is reliable enough, it can be utilized in diabetes diagnosis directly in clinics. Current methods require going on diets and taking blood samples, which could be very time consuming and inconvenient. Using machine learning algorithms is fast and noninvasive compared to the existing methods. The purpose of this research was to build an optimized model by machine learning algorithms that can improve the diagnosis accuracy and classification of patients at high risk of diabetes using retinal imaging.

Download Full-text

Sistem Klasifikasi Tingkat Keparahan Retinopati Diabetik Menggunakan Support Vector Machine

IJEIS (Indonesian Journal of Electronics and Instrumentation Systems) ◽

10.22146/ijeis.31206 ◽

2018 ◽

Vol 8 (1) ◽

pp. 37 ◽

Cited By ~ 1

Author(s):

Taufiq Galang Adi Putranto ◽

Ika Candradewi

Keyword(s):

Support Vector Machine ◽

Diabetic Retinopathy ◽

Correlation Energy ◽

Research Result ◽

Texture Feature ◽

Support Vector ◽

Machine Method ◽

Vision Disorder ◽

Rbf Kernel ◽

Occurrence Matrix

Diabetic retinopathy is a vision disorder disease that can cause damage to the retina of the eye that will have a direct impact on the disruption of vision of the patient. The diabetic retinopathy phase is classified into four types (normal, mild NPDR, moderate NPDR (Non-Proliferative Diabetic Retinopathy), and severe NPDR). Retinal of eye data of diabetic retinopathy patients treated from the MESSIDOR database. By applying image processing, the retinal image of the eye in extraction using the area features extraction from the detection of exudate, blood vessels, microaneurysms, and texture feature extraction Gray Level Co-occurrence Matrix. The extracted results classified using the Support Vector Machine method with the Radial Basis Function (RBF) kernel. Classification evaluated with these parameters: Accuracy, specificity, and sensitivity.The results of classification show the best value using 6 statistical features ie, contrast, homogeneity, correlation, energy, entropy and inverse difference moment in the direction of 45 degrees with the RBF kernel. The result of classification research system on 240 data training and 60 data testing yields an average accuracy is 95.93%, the value of specificity is 97.29%, and a sensitivity rating is 91.07%. From the research result, using RBF kernel get the best accuracy result than using kernel polynomial or kernel linear.

Download Full-text

Evaluation of Performance of Decision Tree, Support Vector Machine and Probabilistic Neural Network Classifiers in a Mobile Based Diabetes Retinopathy Detection System

Asian Journal of Research in Computer Science ◽

10.9734/ajrcos/2019/v3i430099 ◽

2019 ◽

pp. 1-9

Author(s):

O. D. Fenwa ◽

O. O. Alo ◽

I. O. Omotoso

Keyword(s):

Neural Network ◽

Support Vector Machine ◽

Diabetic Retinopathy ◽

Decision Tree ◽

Medical Condition ◽

Detection System ◽

Probabilistic Neural Network ◽

Rejection Rate ◽

Support Vector ◽

Diabetes Retinopathy

Diabetic Retinopathy (DR) is a medical condition where the retina is damaged because fluid leaks from blood vessels into the retina. Ophthalmologists recognize diabetic retinopathy based on features, such as blood vessel area, exudes, hemorrhages, microaneurysms and texture. Aim: The focus of this paper is to evaluate the performance of Decision Tree (DT), Support Vector Machine (SVM) and Probabilistic Neural Network (PNN) Classifiers in Diabetes Retinopathy Detection. Results: Corresponding results showed SVM has the best classification strength by achieving Recognition Accuracy (RA) of 98.50%, while PNN and DT achieved RA of 97.60% and 89.20% respectively. In terms of False Acceptance Rate (FAR) and False Rejection Rate (FRR), SVM has the least values of 7.21, 8.10 while DT and PNN showed 11.10, 9.30 and 13.21, 10.10 respectively. However, in this paper a Mobile based Diabetes Retinopathy Detection System was developed to make the system available for the masses for early detection of the disease.

Download Full-text