Deteksi Kondisi Organ Pankreas Melalui Iris Mata Menggunakan Jaringan Syaraf Tiruan Metode Perambatan Balik dengan Pencirian Matriks Ko-Okurensi Aras Keabuan

Sari

Iridology is the study of iris structure as a reflection of the organ condition and systems in the body. In this study, the organ which detected is pancreas. To determine the condition of the pancreas through the iris, texture analysis and classification process to distinguish the iris of the eye that contains the condition of normal and abnormal pancreas is needed. The purpose of this study was to detect the condition of the pancreas through the iris using artificial backpropagation neural network with the gray level co-occurrence matrices (GLCM) characteristics. Application for the detection of pancreas conditions was made using Matlab version 7.6 (R2008a). Inputs, which used in the study of the eye image, obtained from expert iridolog with both normal and abnormal conditions of the pancreas. The image is then carried out with iris localization process, ROI-making organ of the pancreas, and GLCM feature extraction. Results of feature extraction is used as input data (training data and test data) for the artificial backpropagation neural network method is then used to diagnose pancreatic organ conditions, ie normal or abnormal. GLCM features extraction based on testing for each characteristic texture are average, contrast, correlation, energy, entropy, and homogeneity for the group of normal training data which are valued of 3,217389, 0,233666, 0,632259, 0,575947, 1,379171, and 0,888469 respectively, while for the group of abnormal training data are 0,960503, 0,476226, 0,765723, 0,412549, 2,145339, and 0,824047 respectively. Based on the results of the testing training data, the program can make the correct diagnosis on the incoming data with a success percentage of 95,8%. While based on test results of test data, the program can make the correct diagnosis on the incoming data with a success percentage of 75%.

Keyword : iridology, pancreas, GLCM, artificial neural network, backpropagation.