Effect of different artifact reduction methods for quantitative computed tomography measurements of bone mineral density: a pilot study

The increasing use of metal internal fixation devices has heightened clinical concerns about peri-implant complications, such as osteoporosis, fracture risk, and hardware loosening. Quantitative computed tomography (QCT) is employed for the quantitative assessment of bone mineral density (BMD) at specific anatomical sites in clinical settings, while conventional CT is used to visualize the structures surrounding the implant. However, significant metal artifacts present in the images can impede the observation and diagnosis of adjacent bone structures. This study systematically evaluates the combined use of orthopedic metal artifact reduction (O-MAR) algorithms and virtual monoenergetic imaging (VMI) to improve image quality and the accuracy of BMD measurements within QCT protocols. Three metal screws were inserted into each of five porcine femoral specimens, followed by dual-layer spectral detector CT scanning. Reconstructed images were divided into five groups: conventional imaging (CI), O-MAR, VMI, VMI + O-MAR, and CI images after metal implant removal (CI*). Quantitative analysis was conducted by recording attenuation (HU), calculating the signal-to-noise ratio (SNR) and artifact index (AI) of cancellous bone, and measuring BMD via QCT. Significant differences in HU, SNR, and AI were observed among metal artifact reduction methods. The CI group demonstrated the highest attenuation and SNR (P < 0.05 vs. other groups), while VMI and VMI + O-MAR showed reduced HU values and the lowest SNR (P < 0.05). Both VMI groups exhibited superior artifact reduction (lower AI) compared to other groups. QCT indicated that BMD measurements were lower in the VMI groups compared to the CI or CI* groups (P < 0.05), though O-MAR maintained BMD accuracy comparable to CI*. O-MAR and VMI effectively reduced metal artifacts, although excessive suppression of artifacts by high-energy VMI may compromise bone quality assessment. The standalone application of O-MAR may facilitate more accurate BMD quantification.