An algorithm to calculate physical density of biomedical objects having 3 ≤ Z_eff ≤ 20: an application study using...

Recently, photon-counting computed tomography (PC-CT) devices have become clinically available for X-ray diagnosis, and developments of analysis algorithms to generate various quantitative images have focused attention. This study proposes an algorithm to calculate the effective physical density (ρ_eff) of biological objects having an effective atomic number (Z_eff) of 3-20. In our procedure, ρ_eff was determined when fitting the interaction cross sections (σs) to the measured linear attenuation coefficients (μs), which were determined from virtual monochromatic images (VMIs). In this process, the Z_eff was analyzed in advance, and this information was fed back to the ρ_eff analysis. A key feature of our procedure is that it does not require calibration processes based on specific substances, which allow us to analyze any substance without making assumptions. To validate the availability of the proposed algorithm, we conducted an experiment using a clinical PC-CT scanner. A multi-energy CT phantom, a water phantom, an in-house low-density phantom, food samples, and a chest phantom were scanned. The results showed that our procedure can calculate ρ_eff of water with an uncertainty of 4.5%. The academic significance of the present research results lies in demonstrating that the factors contributing to contrast, previously obtained from CT values, can be physically interpreted using information from Z_eff and ρ_eff. This analysis cannot be performed with conventional single-energy CT scanners and is possible with PC-CT and dual-energy CT scanners. Our novel procedure is expected to provide useful additional information in X-ray diagnosis.