L. Anifah, Haryanto
Osteoarthritis is one of an incurable disease. So far, to reduce the patient's pain, treatment is usually carried out and pain-reducing drugs are given. The purpose of this research is to propose a method of knee osteoarthritis by implementing supervised learning Self Organizing Map (SOM) based on the Junction Space Width (JSW) data, at JSW (0.150) to JSW (0.300). The stages in this study are learning data based on Junction Space Width (JSW) data, minimum JSW medial compartment (mJSW) and JSW fixed location (JSW (x)) at JSW (0.150) to JSW (0.300) using SOM, and to get the performance of the system testing process is carried out. There are 799 data has observed contain: 724 testing data, 75 training data (15 data for normal condition, and 15 data for each the KL-grade. The steps of the research are initialization, determination of cluster number, determine the value of learning rate and iteration value, then calculate the minimum distance between the data and input weight. The experimental results show that using number of iteration 1500 and α0=0.6 that afford system accuracy 59.20%, Grade 0 28.125%, first grade 22.58%, second grade 43.82%, third 89.33 %, and fourth 97.33% are recommended. © 2024, International Organization on 'Technical and Physical Problems of Engineering'. All rights reserved.
Department of Electrical Engineering, Faculty of Engineering, Surabaya State University, Surabaya, Indonesia
