Automated glaucoma detection through machine learning and deep learning Negi Tanishq1, Vashist Apurva2 1Ambedkar DSEU, Shakarpur Campus - I, New-Delhi, India, tn1829@dseu.ac.in 2BCA Department, Ambedkar DSEU, Shakarpur Campus - I, New-Delhi, India, apurva.vashist@gmail.com Online published on 20 August, 2024. Abstract This research delves into the fusion of machine learning (ML) and deep learning (DL) methodologies for the early detection of glaucoma, acknowledging its status as the primary cause of irreversible blindness. Through the utilization of diverse imaging modalities such as optical coherence tomography (OCT) scans and fundus photographs, these algorithms demonstrate proficiency in identifying crucial features indicative of glaucomatous changes. The paper evaluates techniques including support vector machines, random forests, and convolutional neural networks, emphasizing their effectiveness in analyzing data pertaining to the eye. Additionally, it addresses challenges such as class imbalance and data heterogeneity, emphasizing the importance of extensive datasets and transfer learning. Ultimately, the study underscores the necessity of early glaucoma detection and explores the potential of ML and DL techniques in achieving this aim. Top Keywords Glaucoma, Machine learning, Deep learning, Diagnosis, Medical imaging, Convolutional neural networks. Top |