The classification of optical coherence tomography images using machine learning for macular hole and diabetic macular edema diagnoses

Optical Coherence Tomography is a non-invasive medical imaging method used to create images of the retina in eyes, which can be vital to diagnose visual health and diseases. Retinal diseases can partially damage the retina, causing visual problem and even blindness. Therefore, early and precise detection of retinal diseases is crucial to avoid vision deterioration. In recent years, artificial intelligence algorithms have been increasingly developed to automatic identification and classification of medical images that help doctors to improve the accuracy of disease diagnosis. In this work, we propose a simple and effective model of automatic classification for Macular hole and Diabetic macular edema diseases based on optical coherence tomography images by using the Support Vector Machine. Our proposed model is also evaluated by different training data and real images. The high accuracy values of macular hole and diabetic macular edema classifications are up to 95.8% and 86.7% respectively. Our obtained results can be used to effectively assist doctors in the early diagnosis and treatment of such retinal diseases.