A novel metal artifact reducing method for cone-beam CT based on three approximately orthogonal projections

Area of Science:

• Medical Imaging
• Image Reconstruction
• Cone-Beam CT

Background:

• Metal artifacts pose a significant challenge in computed tomography (CT) reconstruction.
• Strictly limited radiation doses in medical CT exacerbate artifact reduction difficulties.

Purpose of the Study:

• To present a novel method for reducing metal artifacts in cone-beam CT (CBCT).
• To improve the precision of metal area segmentation in 2D projections.

Main Methods:

• Calculating 3D coordinates of metallic implants using three nearly orthogonal projection images (one synthesized parallel-beam, two cone-beam).
• Projecting 3D coordinates as metal seed points (MSPs) onto projection images.
• Utilizing region growing segmentation and bilinear interpolation to modify metal areas.

Main Results:

• Accurate localization of metal seed points (MSPs) was achieved.
• The proposed method demonstrated efficiency in reducing metal artifacts.
• Improved segmentation of metal areas was observed.

Conclusions:

• The novel method effectively reduces metal artifacts in CBCT.
• Accurate 3D localization of metallic implants is crucial for artifact reduction.
• This technique offers a promising solution for metal artifact challenges in limited-dose medical imaging.