An application of neural network in medical images

In medical field, doctors often use Greulich&Pyle atlas to determine the skeletal age of a hand radiograph. But some research results have shown that Greulich&Pyle atlas published about 50 years ago may not still be suitable for current population. Meanwhile the results of less experienced doctors may not be reliable and accurate and the results of experienced doctors sometimes are not always available. Here we propose a neural network method based on some previous work and give better results. In this paper, we do some pre-processing on the input data to remove some noise before putting them into the network and some post-processing before outputting the results. Different neural networks such as back-propagation, radias basis network with different architecture are tested. We choose the one with the best performance among them. From the experiments we can see that the results of the neural network are similar to those given by the experienced doctors and better than those of previous research, indicating that this approach is very practical and beneficial to doctors comparing with some other methods currently existing.