EARLY GLAUCOMA DISEASE DETECTION SOFTWARE BASED ON DEEP LEARNING AND FUNDUS IMAGE SEGMENTATION

• Main Author: Javier Kurniawan, Jessen
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/84866
• Institution: Institut Teknologi Bandung

Glaucoma, often referred to as the "silent thief of sight," causes gradual and asymptomatic vision loss, typically becoming noticeable only after significant peripheral vision has already been lost. To diagnose glaucoma, medical professionals employ various methods, including tonometry to measure eye pressure and retinal imaging techniques such as funduscopy. Retinal imaging, a cornerstone of clinical care for patients with retinal diseases, allows for detailed observation of the retina's layers and texture. In the context of glaucoma, fundus images are extensively used for screening to detect damage to the optic nerve head (ONH) located in front of the retina. Technological advancements play a crucial role in enhancing medical diagnostics. Notably, the development of artificial intelligence in the medical field holds significant potential for supporting diagnostic processes. This research aims to develop a mobile application as an alternative technology to facilitate early detection and aid in the diagnosis of glaucoma. The study involves creating and testing an early detection method, implementing it within mobile software, and ensuring the software is user-friendly for both the general public and medical professionals.