Volume Rendering of Tendon--Bone Relationships Using Unenhanced CT

HOME

HELP

FEEDBACK

SUBSCRIPTIONS

Volume Rendering of Tendon—Bone Relationships Using Unenhanced CT

Jason S. Pelc¹ and Christopher F. Beaulieu

^¹ Both authors: Department of Radiology, Stanford University Medical Center, MC 5105, 300 Pasteur Dr., Rm. S-056, Stanford, CA 94305.

View larger version (101K):

[in a new window]

Fig. 1. —Key elements for creating high-quality three-dimensional (3D) volume renderings of muscle—tendon—bone relationships. Top shows axial CT source image (width/level, 800/200 H) from distal forearm of 23-year-old man. "Voxel histogram" is shown on bottom. This represents distribution of voxels (in the entire 3D data set) at each attenuation value. Range of 256 (28) "voxel values" on abscissa shows remapping from original 4096 (2¹²) H scale, in which original histogram was truncated between -200 and +1024 H. This remapping allows finer control over opacity curve than was possible with original 12-bit gray-scale range and allows use of VoxelView's "fast" lighting model. Color scale above abscissa shows how color is mapped to voxel values. Superimposed on voxel histogram (blue) is custom opacity function we designed (yellow graph). Higher values of opacity make range of voxel values more visible, whereas voxels with zero opacity (e.g., fat and air) are rendered transparent.

View larger version (90K):

[in a new window]

Fig. 2A. —Single helical CT acquisition (with 1-mm collimation, 12-cm field of view, 0.5-mm reconstruction interval) of 23-year-old man imaged to characterize fracture at base of second metacarpal (not shown). Three-dimensional image was created from source images with "standard" kernel. Note that tendons (T) appear as smooth continuous structures, that muscles in hypothenar eminence (M) are visible but rendered relatively transparent, and that bone is opaque.

View larger version (106K):

[in a new window]

Fig. 2B. —Single helical CT acquisition (with 1-mm collimation, 12-cm field of view, 0.5-mm reconstruction interval) of 23-year-old man imaged to characterize fracture at base of second metacarpal (not shown). Three-dimensional image displaying same anatomy as A was created from source images with "bone" kernel. Because relatively higher image noise is present than that from standard algorithm, distinction between small attenuation differences is compromised. This problem is illustrated by observation that overall image appears grainy, tendons appear irregular, and muscles cannot be discerned as discrete structures.

View larger version (92K):

[in a new window]

Fig. 3. —21-year-old woman imaged to evaluate talar dome osteochondral lesion (not shown). Similar concepts of voxel opacity mapping and application of custom color table were applied to images reconstructed with "standard" kernel. Note excellent depiction of tendons in ankle, including anterior tibials (AT), posterior tibialis (PT), flexor digitorum longus (FDL), flexor hallucis longus (FHL), and achilles (A). Medial muscle groups (M) are also visible. Here, somewhat higher opacity was applied to muscle attenuation range than to that shown on wrist images shown in Figure 2A,2B. Curvilinear structures representing vessels are seen in several areas (white and black arrowheads) because these structures have similar attenuation to muscle and are surrounded by normal fat, which itself is rendered transparent.

CiteULike

Complore

Connotea

Del.icio.us Add to Digg

Digg

Technorati What's this?