Ocular Toxoplasmosis is caused by the Toxoplasma gondii parasite transmitted by ingestion of raw or undercooked meat, contaminated food or water or by congenital transmission. The inflammation and necrosis of the ocular tissue caused by the infection can reduce visual acuity and even a total loss of vision in the affected eye. The aim of the study is to design a computational learning model based on optimization techniques that reduces computational cost while maintaining optimal diagnostic accuracy. The model is developed in the TensorFlow Lite environment applying quantization and pruning techniques, yielding 0.9843, 0.9545, 0.99 and 0.9767 for accuracy, precision, recall and F1-Score, respectively. This complementary non-invasive diagnostic technique makes it possible to standardize diagnoses and enable early detection of the disease.