Pattern recognition is widely used in various fields of engineering and life-sciences; out of them the fields of medical sciences, bio-informatics and genetics are incorporating the modern and advanced techniques of Pattern Recognition, Machine Learning using classifiers such as Neural Network. The aim of this paper is to present a medical diagnosis oriented neural network based model using two datasets; the Pima Diabetes and Heart Disease. Diabetes Diagnosis is implemented using artificial neural network. The neural network used for the datasets is General Regression Neural Network. Using this neural network, the authors tested the ability of this network to forecast the onset of diabetes mellitus and heart disease.