Advancements in Medical Imaging for Sugar Diagnosis Using Modified Hopfield Neural Network

In the field of medical science, the ability to diagnose diseases from medical images has gained significant importance. Detecting abnormalities in retinal images is particularly challenging and requires specialized expertise.The following paper proposes a Modified Hopfield Neural Network (MHNN) for accurately diagnosing Diabetic Retinopathy from retinal images. The MHNN adjusts both weight and output values simultaneously, unlike conventional neural networks, leading to improved accuracy. The proposed method was tested on 540 images and achieved an average sensitivity and specificity of 0.99 and accuracy of 99.25%. Compared to other neural network approaches, the proposed MHNN is superior in diagnosing Diabetic Retinopathy from Retinal Images.