EFFICIENT RETINAL IMAGE SEGMENTATION BY U-NET FOR AGE-RELATED MACULAR DEGENERATION DIAGNOSIS

Age-related Macular Degeneration (AMD) is a prominent factor contributing to visual impairment in older people, characterized by deterioration of the macula, the center part of the retina. An accurate segmentation of blood vessels is essential for successful intervention and control of MAD. This study proposes an approach for effectively segmenting blood vessels using U-Net architecture. It is a specialized convolutional neural network that has shown significant potential in accurately segmenting intricate structures captured in medical images. It uses U-Net to precisely define the blood vessels from retinal images, facilitating accurate identification of macula regions for early AMD detection. The efficacy of the proposed method in achieving high accuracy and the computational economy is shown by its evaluation of a large dataset, Structured Analysis of the Retina (STARE). The findings demonstrate that the U-Net-based approach outperforms existing segmentation methods in accuracy and efficiency, making it a promising tool for identifying and monitoring AMD.