Author ORCID Identifier


Date of Graduation


Document Type

Dissertation (PhD)

Program Affiliation

Medical Physics

Degree Name

Doctor of Philosophy (PhD)

Advisor/Committee Chair

David Followill

Committee Member

Stephen Kry

Committee Member

Paige Taylor

Committee Member

Xiaodong Zhang

Committee Member

Steven Frank

Committee Member

John Rong


Purpose: Dental amalgams (high Z materials) are common sources of artifacts in Head and Neck (HN) images. Commercial artifact reduction techniques have been offered, but many are impractical, produce inaccurate CT images or are not clinically available, thus not widely implemented. The goal of this work is to use CT gantry tilts to develop and evaluate a stereoscopic HN metal artifact management algorithm and investigate its improvement in proton treatment planning.

Methods: The in-house CT metal artifact management method for proton planning (AMPP) uses two angled CT scans to generate a single image set with no metal artifacts posterior to the dental metal implants. The algorithm was evaluated (geometrical distortion and HU accuracy) using a geometrical phantom simulating a HN patient with dental fillings. A HN anthropomorphic phantom composed of proton tissue equivalent materials, human skull, air cavities was used to perform a quantitative image quality comparison between AMPP and four major CT vendors’ commercial metal artifact reduction (MAR) solutions (OMAR from Philips, iMAR from Siemens, SEMAR from Canon, SmartMAR from GE), along with their implications on proton dose distributions.

Results: The in-house algorithm designed produced geometrically and HU accurate images free of metal artifacts posterior to the oral cavity. AMPP outperformed all vendors’ solutions in terms of image quality, showing lower HU differences and fewer bad pixels (4.2% compared to 25.5-65.5%). Dose distributions were negatively impacted by the presence of metal artifacts; the vendor solutions provided varying, but suboptimal, mitigation of this effect. Our in-house algorithm (AMPP) outperformed the vendor’s solutions on all treatment plans and showed the most comparable DVHs to the baseline (no metal).

Conclusion: A novel in-house algorithm was designed that produces geometrically and HU accurate images free of CT metal artifacts posterior to the HN region. Commercial MAR algorithms were ineffective at reducing artifacts in a HN geometrical and anthropomorphic phantom scenario. Correspondingly, they were not successful at mitigating the impact of artifacts on proton dose distributions. Our in-house algorithm outperformed all four commercial vendor solutions in both imaging and dose distributions, and is ready to be implemented on patients.


CT, Tilted angle, Algorithm, metal artifact reduction, MAR, image quality, treatment planning, proton dosimetry, dose calculation, stereoscopic



To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.