A NEW NEURAL-NETWORK STRUCTURE FOR DETECTION OF CORONARY HEART-DISEASE

We investigate the application of neural network for the detection of Coronary Heart Disease (CHD). We have used a Neural Network (NN) on da ta from a self-applied questionnaire to implement a decision system de signed to seek out high risk individuals in a large population. A Mult i-Layered Perceptron (MLP) was trained with risk factors to distinguis h CHD. We also describe a modification to the architecture of the neur al network in which an extra layer of neurons is added at the input. W e present possible interpretations of the weights of these neurons, an d show how they can be used as a selection criteria for which question s to use as inputs. The technique is compared against other statistica l methods. We go on to demonstrate the system's capability for detecti ng both the symptomatic and asymptomatic patient.