Title :
Detection of blood vessels in 3D MR-images
Author :
Grönqvist, Anders ; Lenz, Reiner
Author_Institution :
Dept. of Electr. Eng., Linkoping Univ., Sweden
Abstract :
Rotation-invariant operators in combination with neural nets are shown to be an effective method for detecting complex structures in 3D gray-scale images. Rotation-invariant operators are used as feature extractors to generate input vectors to a neural net. The feature vectors are classified by the neural net, and the result is finally filtered from noise.<>
Keywords :
biomedical NMR; blood; computerised pattern recognition; computerised picture processing; medical computing; neural nets; 3D MR-images; blood vessels; complex structures; feature extractors; gray-scale images; neural nets; noise; pattern recognition; picture processing; rotation-invariant operators; Biomedical computing; Biomedical magnetic resonance imaging; Blood; Image processing; Neural networks; Pattern recognition;
Conference_Titel :
Neural Networks, 1989. IJCNN., International Joint Conference on
Conference_Location :
Washington, DC, USA
DOI :
10.1109/IJCNN.1989.118572
