PREDICTION OF TRAUMA MORTALITY USING A NEURAL-NETWORK

A neural network is a computerized construct consisting of input neuro ns (which process input data) connected to hidden neurons (to mathemat ically manipulate values they receive from all the input neurons) conn ected to output neurons (to output a prediction). Neural networks are created and trained via multiple iterations over data with known resul ts. In 1993, 897 trauma patients were either declared dead in the emer gency room (ER; 76 cases), admitted to the intensive care unit (427 ca ses, 36 deaths), or taken directly to the operating room (394 cases, 2 9 deaths). Using only data available from the ER, a neural network was created, and 628 cases were randomly selected for training. After 268 iterations, the network was trained to correctly predict death or sur vival in all 628 cases. This trained network was then tested on the ot her 269 cases without our providing the death or survival result. Its overall accuracy was 91 per cent (244 of 269 cases). It was able to pr edict correctly 60 per cent (12 of 20 cases) of the postoperative or p ost-intensive care unit admission deaths and 90 per cent (26 of 29 cas es) of the deaths in the ER. Computerized neural networks can accurate ly predict a trauma patient's fate based on inital ER presentation. Th e theory and use of neural networks in predicting clinical outcome wil l be presented.