E. Somoza et Jr. Somoza, A NEURAL-NETWORK APPROACH TO PREDICTING ADMISSION DECISIONS IN A PSYCHIATRIC EMERGENCY ROOM, Medical decision making, 13(4), 1993, pp. 273-280
Clinical decision making is based on recognizing complex patterns of p
atients signs and symptoms. Neural networks have been shown to be very
effective at this type of pattern recognition, and in this study a ne
ural-network approach was used to predict which patients seen in a psy
chiatric emergency room required admission and which did not. Data fro
m all walk-in patients (N = 658) evaluated during normal working hours
in a psychiatric emergency room during a one-year period were used ei
ther to train a neural network or to test its performance. The network
had 53 input nodes, one hidden layer, and an output layer with a sing
le node. The back-propagation method was used to train the network. Th
e neural network's admitting decisions were in substantial agreement w
ith those of the clinicians (kappa coefficient = 0.63). When used as a
diagnostic test for admission it had a specificity of 94%, a sensitiv
ity of 70%, and an overall accuracy of 91%. The information gain was 3
5% of that of a perfect diagnostic test. These results show that a neu
ral network can be trained to make clinical decisions that are in subs
tantial agreement with those of experienced clinicians.