Improved Convolutional Neural Network with BIRCH based Segmentation for Glaucoma Detection

Glaucoma is a second-leading disease in the globe that causes permanent harm to the eye nerves and can cause partial or whole visual impairment. Glaucoma is known as the "silent thief of sight" as it frequently goes undiagnosed in its earliest stages. According to studies, glaucoma is still a major worldwide health problem. The efficacy of these ML techniques may be limited by human specialists' tendency to ignore or misrepresent important details required for precise glaucoma diagnosis. Nowadays, deep learning methods have gained popularity in the healthcare and medical fields. A novel Glaucoma Detection approach using an improved DL technique. Primarily, the fundus images (input) are pre-processedusing a median filter. Then, BIRCH-based model is deployed for segmentation. Further, ResNet, VGG 16 and statistical featuresare extracted. Finally, Improved CNNwith modifications isproposed for the detection of Glaucoma.