I. Introduction

Diabetes risk is a condition correlated with some asymptomatic disease like PCOD and pancreatitis. The lifestyle disorder like stress and steroid leads to insulin resistant PCOD and pancreatitis. Insulin resistant is the underlying condition leads to PCOD and pancreatitis may reflect in retina as myopia and CSR respectively. The neural network shows high prediction accuracy with the given dataset. The input dataset consists of OCT retinal parameters with hormone profile of PCOD and pancreatitis patients. The neural network consists of input, hidden and output layer to process the given data efficiently. Neural network consists of important layers such as input layer, hidden layer and output layer. The given dataset is divided as training and testing dataset to train the model with training dataset. The neural network consists of two different parameters such as network parameters and hyper parameters. The network parameters consist of weight and bias values will be assigned by neural network by itself. The hyper parameters are tuned to improve the performance of the model.