Expert System for Diagnosis and Appropriate Medical Prescription of Heart Disease Using Radial Basis Function

The use of neural network on different diseases has been used on large scale since last two decades. This paper includes details about patients data, coding, normalization and tabulation. The Feed Forward Back-propagation (FFBP) and Radial Basis Function(RBF) has been applied over the data for the experiment. Radial basis function can require more neurons than standard feedforward backpropagation networks, but often they can be designed in a fraction of the time it takes to train standard feedforward networks. Sahara Hospital, Roshan gate, Aurangabad has played an encore role for collecting the information of 375 patients under the supervision of Dr. Abdul Jabbar (MD Medicine). After a thorough clinica l diagnosis the above citied number of patients were found to be suffering from heart diseases. To make the system more authentic and reliable out of 375 patients 250 patients were used for training set and 125 for evaluation process. In conclusion, the experiments proved beneficial as it gave maximum positive result upto 97%. This expert system of clinical diagnosis may be of great use to the trainees as well as the specialist of heart diseases.