METHOD AND DEVICE FOR AUTOMATICALLY DETECTING FEATURE POINTS OF THREE-DIMENSIONAL MEDICAL IMAGE DATA USING DEEP LEARNING, METHOD FOR AUTOMATING POSITION ALIGNMENT OF DENTAL THREE-DIMENSIONAL DATA, METHOD FOR AUTOMATICALLY DETECTING LANDMARKS IN DENTAL THREE-DIMENSIONAL SCAN DATA, METHOD FOR DETERMINING MATCHING ACCURACY OF THREE-DIMENSIONAL DENTAL CT IMAGE AND THREE-DIMENSIONAL DIGITAL IMPRESSION MODEL, AND COMPUTER-READABLE RECORDING MEDIUM IN WHICH PROGRAM FOR EXECUTING METHODS IN COMPUTER IS RECORDED

摘要

A method for automatically detecting feature points of three-dimensional medical image data using deep learning comprises the steps of: receiving input of a three-dimensional medical volume image; generating two-dimensional brightness value projection images on the basis of the three-dimensional medical volume image; automatically detecting initial anatomical feature points using a first convolutional neural network on the basis of the two-dimensional brightness value projection images; generating a three-dimensional volume of interest region on the basis of the initial anatomical feature points; and automatically detecting fine anatomical feature points using a second convolutional neural network, different from the first convolutional neural network, on the basis of the three-dimensional volume of interest region.