Paper Info
Title
Knowledge-Driven Extraction of the Four Modified Talairach Cortical Landmarks (A, P, L, and R) from MR Neuroimages
Abstract
An algorithm to locate the four modified Talairach corticallandmarks (A, P, L, and R) based on knowledge is proposed.Knowledge is employed to select the axial plane (the APplane) passing through the anterior and posteriorcommissures, determination of optimum thresholds,segmentation, and refinement of the AP plane. Thealgorithm has been validated against 38 T1-weighteddatasets from public resources. It takes less than 1 secondon Pentium 4 (2.6 GHz) to extract the 4 landmarks. Theaverage landmark location error is below 1 mm. Thealgorithm is robust and accurate as the factors influencingthe determination of cortical landmarks like noise, graylevel inhomogeneity, partial volume effect, and connectionbetween the cortex and the sagittal sinus/meninges/skull arecarefully compensated. A low computational cost isattributed to simple operations like thresholding, seeding,simple morphological operations and distance transform.
Year
DOI
Venue
2004
10.1109/BIBE.2004.1317330
BIBE
Keywords
Field
DocType
graylevel inhomogeneity,simple operation,simple morphological operation,mr neuroimages,low computational cost,theaverage landmark location error,knowledge-driven extraction,factors influencingthe determination,modified talairach cortical landmarks,cortical landmark,modified talairach corticallandmarks,ap plane,axial plane,image segmentation,neurophysiology,flowcharts,partial volume effect,distance transform,biomedical imaging,segmentation,radio frequency,feature extraction
Computer vision,Transverse plane,Medical imaging,Computer science,Segmentation,Image segmentation,Feature extraction,Distance transform,Artificial intelligence,Thresholding,Partial volume
Conference
ISBN
Citations 
PageRank 
0-7695-2173-8
0
0.34
References 
Authors
9
2
Name
Order
Citations
PageRank
Qingmao Hu116019.73
Wieslaw L. Nowinski233747.85