The Assessment of Time-Domain Features for Detecting Symptoms of Diabetic Retinopathy

Elibol, Gülin; Ergin, Semih

doi:10.18201/ijisae.270351

Cited by 2 publications

(2 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These parameters were used to find a new covariance matrix. In fact, small variance directions are eliminated due to the regulation process and then a new covariance matrix is calculated via the leading main components and the smallest eigenvalues [29].…”

Section: Lbncmentioning

confidence: 99%

Classification of Dynamic Egg Weights Using Feature Extraction Methods

Seçil

Yumurtacı

Ergin

et al. 2020

Eskişehir Technical University Journal of Science and Technology a - Applied Sciences and Engineering

View full text Add to dashboard Cite

In this study, a feature vector is determined in order to classify chicken eggs into four different weight groups by using the dynamic weighing system, and then the success rates of different classifiers in the process of weight classification are analyzed. The class numbers and labels of the egg weight groups are determined as 1-extra large, 2-large, 3-middle, and 4small, respectively. The dynamic weighing system is made of three components; mechanic system, electronic control board, and software. Firstly, a data set is created on the basis of analogue egg weight data obtained from the dynamic weighing system. From the obtained data set, three different feature vectors are extracted by using Time-Domain (TD), Power Spectral Density (PSD) and Discrete Wavelet Transform (DWT) based methods. The extracted feature vectors are then applied to Linear Bayes Normal Classifier (LBNC), Fisher's Linear Discriminant Analysis (FLDA), Support Vector Machine (SVM), Decision Tree (DT), and K-Nearest Neighborhood (k-NN) classifiers respectively and egg weight classes are determined. A five-fold crossvalidation is carried out in order to confidentially test the performance of classification. As can be seen from the experimental results, both feature vectors and classifiers are highly successful in determining the weight classes of eggs. It is observed that the most successful features are the entropy values of DWT with a classification rate of 97.01% for the k-NN classifier.

show abstract

Section: Lbncmentioning

confidence: 99%

Classification of Dynamic Egg Weights Using Feature Extraction Methods

Seçil

Yumurtacı

Ergin

et al. 2020

Eskişehir Technical University Journal of Science and Technology a - Applied Sciences and Engineering

View full text Add to dashboard Cite

show abstract

“…Multi-layer perceptron was found most suitable in predicting risk level as it has the highest accuracy. Elibol and Ergin (2016) extracted time domain features from retina images and those features are used to classify the stage of diabetic retinopathy in which an image is in. The algorithms used are Fisher's linear discriminant analysis, Linear Bayes Normal classifier, Decision tree and k-Nearest Neighbour.…”

Section: Introductionmentioning

confidence: 99%

Application of Data Mining Algorithms for Feature Selection and Prediction of Diabetic Retinopathy

Oladele

Ogundokun

Kayode

et al. 2019

Computational Science and Its Applications – ICCSA 2019

View full text Add to dashboard Cite

Diabetes Retinopathy is a disease which results from a prolonged case of diabetes mellitus and it is the most common cause of loss of vision in man. Data mining algorithms are used in medical and computer fields to find effective ways of forecasting a particular disease. This research was aimed at determining the effect of using feature selection in predicting Diabetes Retinopathy. The dataset used for this study was gotten from diabetes retinopathy Debrecen dataset from the University of California in a form suitable for mining. Feature selection was executed on diabetes retinopathy data then the Implementation of k-Nearest Neighbour, C4.5 decision tree, Multi-layer Perceptron (MLP) and Support Vector Machines was conducted on diabetes retinopathy data with and without feature selection. There was access to the algorithms in terms of accuracy and sensitivity. It is observed from the results that, making use of feature selection on algorithms increases the accuracy as well as the sensitivity of the algorithms considered and it is mostly reflected in the support vector machine algorithm. Making use of feature selection for classification also increases the time taken for the prediction of diabetes retinopathy.

show abstract

The Assessment of Time-Domain Features for Detecting Symptoms of Diabetic Retinopathy

Cited by 2 publications

References 16 publications

Classification of Dynamic Egg Weights Using Feature Extraction Methods

Classification of Dynamic Egg Weights Using Feature Extraction Methods

Application of Data Mining Algorithms for Feature Selection and Prediction of Diabetic Retinopathy

Contact Info

Product

Resources

About