Osteoarthritis Classification Using Self Organizing Map Based Gray Level Run Length Matrices

The number of knee osteoarthritis patients in the world is very high. This disease can not be cured, and can only be treated so that the condition does not get worse. Detecting the status of osteoarthritis early is very important, the patient can be treated according to KL-Grade and hopefully the quality of life of patient will be improved. This study discusses Gray Level Run Length Matrices feature-based classification using Self Organizing Map. The classification results show KL-Grade 0 accuracy is 20.68%, KL-Grade 1 has an accuracy of 25.86%, accuracy for KL-Grade 2 is 20.69%, KL-Grade 3 has an accuracy only 30.17%, and KL-Grade 4 just only 12 % could classified. Based on the results obtained then further research is needed to obtain better results.