Neural networks are a new type of computing algorithm. They are especially useful in pattern recognition. In this study we applied neural networks to the prediction of lymph node metastasis of patients with oral cancer. A data set of 1,116 lymph nodes verified histopathologically was used to train and evaluate the neural networks. Various three-layer feed-forward networks with a back-propagation algorithm were employed in this study. Performance of the neural networks was compared with that of radiologists and discriminant analysis (Quantification theory type II). Neural networks had a sensitivity of 80.6% and a specificity of 94.6%. Diagnostic accuracy of the neural networks was 93.6%, which was comparable to those of discriminant analysis and clinical radiologists.
All Science Journal Classification (ASJC) codes
- Dentistry (miscellaneous)
- Radiology Nuclear Medicine and imaging