Novel multi-scan-body cone-beam CT: comparison with photon-counting and energy-integrating CT in an anthropomorphic hand phantom

Background: Cone-beam computed tomography (CBCT) can offer advantages over multidetector CT in dose efficiency and economic costs, but musculoskeletal applications were limited in gantry-free systems. Purpose To assess the utility of novel multi-scan-body CBCT for osseous imaging, compared to clinically implemented photon-counting-detector (PCCT) and energy-integrating-detector (EICT) CT. Material and Methods: An anthropomorphic hand wrist phantom underwent gantry-based CBCT (low-dose, regular, enhanced, and best settings), PCCT, and EICT. Quantitative metrics included dose values, noise, and noise power spectrum (NPS). Three radiologists with varying experience levels (10, 6, and 1 years) assessed depiction of cortical and trabecular bone, articular surfaces, intraosseous ganglion cyst, and overall image quality using 5-point Likert scales. Results: Low-dose and regular CBCT (0.37 and 0.67 mGy) showed the lowest dose values (CTDI $_{vol}$ ), followed by EICT, enhanced and best CBCT, and then PCCT (0.76, 1.08, and 1.61, and 3.56 mGy, respectively). Absolute noise was lowest for PCCT (15.1), followed by best (23.2), regular (25.1), and enhanced (27.4) CBCT. Highest noise was measured for low-dose CBCT (35.1) and EICT (30.1). CBCT showed overall irregular and relatively high NPS, compared to regular and high NPS of EID, whereas PCCT showed a cleaner texture with the lowest NPS. Qualitatively, CBCT (enhanced, best) generally achieved the best scores, while the other scans scored equally well. Average interreader agreement ranged from moderate to near-perfect (k = 0.53–0.87). Conclusion: Novel multi-scan-body CBCT with variable image quality settings can provide detailed depiction of fine osseous structures, demonstrating comparable or lower doses compared to clinically implemented PCCT and EICT.