The use of data augmentation in medical images has become an effective technique to improve the performance of deep learning models, especially when dealing with a limited dataset. This technique involves applying transformations such as rotations, scaling, and brightness adjustments to the original images, which allows for the generation of artificial variations and thus increases the diversity of the dataset. This study investigates the impact of data augmentation on retinal images affected by Age-Related Macular Degeneration (AMD), one of the leading causes of vision loss in older adults. To achieve this, the Swin Transformer V2 model was used, a deep learning model based on the transformer architecture, known for its ability to capture spatial patterns in high-resolution images. Various data augmentation techniques were applied, evaluating their effect on the accuracy and generalization ability of the models trained with Swin Transformer V2 for AMD detection. The results show that the use of these techniques not only improves the accuracy of the models but also optimizes their ability to generalize, achieving an accuracy of 88.64%, which is crucial for the early detection of this disease. In this way, the study emphasizes the value of data augmentation in computer-assisted diagnosis in the field of visual health, contributing to better clinical management of AMD and the improvement of diagnostic methods in modern medicine.