A quantitative analysis of metal artifact reduction algorithm performance in volume correction with 3 CBCT devices
Karla de Faria Vasconcelos1, Polyane Mazucatto Queiroz2, Marina Codari3
1OMFS IMPATH research group, Department of Imaging and Pathology, Faculty of Medicine, University of Leuven and Oral & Maxillofacial Surgery, University Hospitals Leuven, Leuven, Belgium; Division of Oral Radiology, Department of Oral Diagnosis, School of Dentistry of Piracicaba, University of Campinas, Piracicaba, Sao Paulo, Brazil.
Metal artifact reduction (MAR) algorithms significantly altered metal cylinder volumes, particularly with amalgam and copper-aluminum (CuAl) alloys. Performance varied by cone beam computed tomography (CBCT) device, highlighting material and scanner dependency.
Area of Science:
- Dental imaging
- Medical physics
- Radiology
Background:
- Metal artifacts in cone beam computed tomography (CBCT) can distort image accuracy.
- Metal artifact reduction (MAR) algorithms aim to mitigate these distortions.
- Understanding MAR performance across different materials and devices is crucial for diagnostic reliability.
Purpose of the Study:
- To quantitatively evaluate the effectiveness of MAR algorithms in correcting the volume of metal artifacts.
- To assess the influence of different metal materials (amalgam, CuAl, titanium), positions, and fields of view (FOVs) on MAR performance.
- To compare MAR performance across three distinct CBCT scanners.
Main Methods:
- Nine phantoms with amalgam, CuAl, and titanium cylinders were scanned using three CBCT devices (NewTom VGi evo, Picasso Trio, ProMax 3-D Max) with two FOVs.
- MATLAB software analyzed volume differences before and after MAR application.
- Statistical analysis included Wilcoxon's and Kruskal-Wallis tests (significance level 5%).
Main Results:
- MAR significantly impacted the volumes of amalgam and CuAl cylinders.
- No significant volume differences were observed post-MAR concerning cylinder positions or FOVs.
- MAR effectiveness on cylinder volumes was scanner-dependent, noted in NewTom VGi evo and ProMax 3-D Max, but not Picasso Trio.
Conclusions:
- The efficacy of MAR algorithms in correcting metal object volumes is contingent upon the specific metal material.
- The performance of MAR algorithms is also dependent on the characteristics of the CBCT imaging unit used.
