Artificial intelligence is already part of our lives, look at videogames that respond differently depending on what the user does, cellphones that respond to voice commands or even your email that is able to detect spam. These technologies use different learning algorithms and almost all of them use machine learning and pattern recognition techniques. The current methods of classification are part of these techniques and give us the possibility to construct machines capable of learning from experience. This work is centered in understanding the fundamental theory of some of the most common and used classification methods, such as Nearest Neighbors, Decision Trees, Support Vector Machines and Neural Networks. These methods are then used to solve a medical diagnose problem: classification of the degree of Diabetic Retinopathy (DR) disease in back-of-the-eye images. DR is a condition where the retina is damaged due to fluid leaking from the blood vessels into the retina and is the major cause of blindness in the world. It has been shown that early diagnosis can play a major role in prevention of visual loss and blindness. This is the main reason for which a correct classification between the different stages of the DR has been studied and attempts are made to develop automatic classification. In this text a comparison and analysis of the classification methods is made with the objective of solving this classification problem and determine which of the classifiers considered works best for this problem.