Premium
Automatic detection of fiducial markers in fluoroscopy images for on‐line calibration
Author(s) -
Smith Laurence,
Pleasance Mike,
Seeton Rosalyn,
Archip Neculai,
Rohling Robert
Publication year - 2005
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1118/1.1914547
Subject(s) - fiducial marker , fluoroscopy , calibration , artificial intelligence , medical imaging , computer vision , computer science , line (geometry) , standard deviation , nuclear medicine , mathematics , radiology , medicine , geometry , statistics
Calibrated C‐arm fluoroscopy is used for a variety of medical procedures where objects and anatomical structures need to be located in space. Calibration is often based on imaging a grid of fiducial markers and using the C‐arm image's geometrical measurements (radius and center) together with the positions of the markers. An on‐line technique is developed to automatically locate the fiducial markers and validated on 97 images. The success rate of the detection algorithm is 96.28% with an average error of 0.46 mm and a standard deviation of 0.32 mm .