A neural network to predict symptomatic lung injury

A nonlinear neural network that simultaneously uses pre-radiotherapy (RT) b iological and physical data was developed to predict symptomatic lung injur y. The input data were pre-RT pulmonary function, three-dimensional treatme nt plan doses and demographics. The output was a single value between 0 (as ymptomatic) and 1 (symptomatic) to predict the likelihood that a particular patient would become symptomatic. The network was trained on data from 97 patients for 400 iterations with the goal to minimize the mean-squared erro r. Statistical analysis was performed on the resulting network to determine the model's accuracy. Results from the neural network were compared with t hose given by traditional linear discriminate analysis and the dose-volume histogram reduction (DVHR) scheme of Kutcher. Receiver-operator characteris tic (ROC) analysis was performed on the resulting network which had Az = 0. 833 +/- 0.04. (Az is the area under the ROC curve.) Linear discriminate mul tivariate analysis yielded an Az = 0.813 +/- 0.06. The DVHR method had Az = 0.521 +/- 0.08. The network was also used to rank the significance of the input variables. Future studies will be conducted to improve network accura cy and to include functional imaging data.