Optic cup and optic disc segmentation using improved selfish gene algorithm / Norharyati Md Ariff

Glaucoma is a disease that is defined by the pressure increased with the eyeball, causing severe damage to the optic nerve. One of the symptoms in glaucoma disease detection is an increased fluid pressure, which in the long term will damage the eye's optic nerve and it may in the worst case lead to blindness. Blindness due to optic nerve damage is irreversible unless it is intervened with proper treatment. In view of this, eye screening is important for early detection. Currently, the very important indicator for accessing the progression of glaucoma is the cup-to-disc ratio (CDR). Due to the complexity of Cup to Disc Ratio (CDR) measurement where the visibility of the boundary between optic cup and optic disc with high density vascular in the optic region, this research explores the methods that can detect the optic cup and optic disc by using digital fundus image as a cheaper solution for an eye screening. Image processing techniques were employed to segment and extract the optic cup and optic disc for glaucoma detection purpose. This study performed using a new bio-inspired algorithm; Selfish Gene Algorithm (SFGA) for optic cup and optic disc segmentation. In addition, this new algorithm is compared to color channel multi-thresholding segmentation and artificial intelligence segmentation based clustering method such as Adaptive Neuro-Fuzzy Inference System (ANFIS) and Fuzzy cMeans (FCM). From the results and analysis obtained from this research, it is established that improved SFGA is outperformed ANFIS, FCM, and Color Channel Multi-thresholding. Therefore, SFGA has potential to greatly improve outcomes for the current technology.