CPC G06T 19/00 (2013.01) [G06T 5/70 (2024.01); G06T 7/0012 (2013.01); G06T 7/73 (2017.01); G06T 2200/04 (2013.01); G06T 2207/10081 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30012 (2013.01); G06T 2219/004 (2013.01)] | 16 Claims |
1. A method for positioning vertebra in an CT image, the method comprising following steps:
pre-processing vertebra CT image data;
inputting the pre-processed vertebra CT image data into a pre-trained neural network to obtain regression results of heat maps of key points corresponding to the pre-processed vertebra CT image data;
regressing of 3D heat maps corresponding to the positions of the key points of the vertebra mass center based on the regression results of the heat maps of the key points and the pre-processed vertebra CT image data; and
serving 3D heat maps corresponding to the positions of the key points of the vertebra mass center as labels, and networked regressing 3D heat map information to position the vertebra; wherein the step of inputting the pre-processed vertebra CT image data into a pre-trained neural network to obtain regression results of heat maps of key points corresponding to the pre-processed vertebra CT image data comprising:
inputting the pre-processed vertebra CT image data into the pre-trained neural network with the high resolution ratio; the neural network with the high resolution ratio includes feature images with different resolution ratios being parallel processed; and
interactive processing the pre-processed vertebra CT image data with the feature images with different resolution ratios, and serving 25 3D heat maps as network labels to obtain regression results of the heat maps of 25 key points corresponding to the vertebra.
|