Prediction of Coronary Artery Disease from Iris Images Using Local Binary Patterns and Artificial Neural Network
Coronary Artery Disease (CAD) is a heart disease that occurs as a result of narrowing or occlusion of the coronary arteries that feed the heart muscle. Early diagnosis of CAD, which is a health problem with a high mortality rate worldwide, is very important. In this study, it was aimed to predict HR using iridology and image processing techniques. Unlike the existing studies, the performance of the Local Binary Patterns (YBP) feature extraction method, which was not used in heart disease prediction studies performed with iridology, was analyzed.
In the proposed method, features were extracted from the iris images of a total of 198 volunteers, 94 of whom were in the CAD and 104 in the Control group, and the classification was performed using the Artificial Neural Network (ANN). The Integral Differential Operator was used to find the iris positions in the image and the Rubber Sheet Normalization methods were used to convert the iris to rectangular format. By means of the iridology map, the heart region in the iris was determined as the analysis region, and 59 histogram-based features were extracted from this region with one pixel and eight neighborhoods with the IRS. The extracted features were classified by ANN. In the data divided into two groups as training and test, the training process was carried out with the Scaled Conjugate Gradient (SCG) algorithm. The accuracy, precision, sensitivity, specificity, F1 score and Area Under the Curve (AUC) values determined as performance criteria were 91.5%, 0.9063, 0.9355, 0.8929, 0 for the test data, respectively. 0.92063 and 0.9103 were found. In line with the findings obtained, it can be said that the proposed method based on FAI is successful in estimating CAD.