Evaluation of a metal artifacts reduction algorithm applied to postinterventional flat panel detector CT imaging.

Area of Science:

• Medical Imaging
• Radiology
• Image Processing

Background:

• Flat panel detector CT (FDCT) imaging is susceptible to streak artifacts from radiodense metallic implants.
• These artifacts degrade image quality, particularly in neurovascular applications.

Purpose of the Study:

• To evaluate a novel metal artifacts reduction (MAR) prototype algorithm for FDCT.
• To assess the algorithm's efficacy in a routine clinical setting for neuroimaging.

Main Methods:

• 59 patients undergoing or previously having undergone cerebral endovascular/surgical procedures were included.
• Physicians rated artifact severity adjacent to and 3 cm from implants on a 3-point scale.
• Visible vessel segments were counted within a 3-cm radius before and after MAR application.

Main Results:

• The MAR algorithm was applied successfully to all 59 FDCT datasets.
• Significant reductions in metal artifacts were observed both immediately adjacent to (P = .05) and 3 cm from (P = .03) implants.
• A trend towards increased visualization of vessel segments was noted (4.07 to 5.29, P = .1235).

Conclusions:

• Metal artifacts reduction is effective in improving FDCT images degraded by metallic implants.
• The MAR algorithm significantly decreases artifacts and shows a trend toward enhanced vessel visualization.