ABSTRACT
Glaucoma is a progressive optic neuropathy characterized by the degeneration of retinal ganglion cells and their axons, leading to vision loss.
The early detection and treatment of glaucoma are crucial for preventing visual impairment. However, glaucoma is often asymptomatic in its early stages, and traditional methods for detecting glaucoma, such as tonometry and visual field testing, are invasive, time-consuming, and subjective.
Transfer learning, a machine learning technique that involves the use of pre-trained models on one task as the starting point for training on a new task, has recently been applied to the detection of glaucoma.
In transfer learning, the pre-trained model serves as a feature extractor, and the task-specific layers are trained on the new data.
This approach has been effective in reducing the amount of annotated data required for training and in improving the performance of the model.
The main aim of this work was to explore “EfficentNetV2” series transfer learning models and the variety of datasets related to glaucoma in different combinations to identify the models that are best suited for glaucoma detection.
