Advanced Iterative Metal Artifact Reduction With Intrinsic Spectral Information From Photon-counting Detector...

Area of Science:

• Medical Imaging
• Radiology
• Biomedical Engineering

Background:

• Metallic implants in computed tomography (CT) scans cause significant artifacts, degrading image quality and diagnostic accuracy.
• Existing metal artifact reduction (MAR) techniques show limitations, especially in areas close to implants.

Purpose of the Study:

• To evaluate a novel iterative metal artifact reduction (iMAR) algorithm for photon-counting detector CT (PCD-CT).
• Focus on visualizing the critical metal-bone interface around a fixation nail implant.

Main Methods:

• Scanned bovine femurs with titanium implants on a PCD-CT scanner.
• Reconstructed images with no iMAR, current iMAR, and a new prototype iMAR algorithm utilizing spectral information.
• Quantified artifact severity and assessed qualitative image quality and diagnostic confidence by radiologists.

Main Results:

• The new iMAR algorithm significantly reduced artifact severity in bone adjacent to the implant (159 HU) compared to no iMAR (226 HU) and current iMAR (174 HU).
• Artifacts adjacent to the femur were also reduced by the new iMAR (29 HU).
• New iMAR achieved higher qualitative scores and avoided newly introduced artifacts seen with the current iMAR.

Conclusions:

• The advanced iMAR algorithm effectively reduces metal artifacts in PCD-CT imaging.
• Improved visualization of the metal-bone interface is achieved, with potential for enhanced assessment of implant-related issues like aseptic loosening.