[Automated cephalometric landmark identification and location based on convolutional neural network].

The model proposed in this paper could adapt to the absence of landmark in lateral cephalograms and locate 61 commonly used landmarks with high accuracy to meet the requirements of different cephalometric analysis methods.