Quadratic relation for mass density calibration in human body using dual-energy CT data

Area of Science:

• Medical Imaging
• Biophysics
• Computational Physics

Background:

• Dual-energy computed tomography (DECT) offers advanced tissue characterization.
• Accurate mass density (ρ) estimation is crucial for quantitative imaging.
• Existing methods for deriving ρ from DECT data often require complex tissue segmentation.

Purpose of the Study:

• To develop and validate a novel method, DEEDZ-MD, for deriving human tissue mass density (ρ) from DECT data.
• To calibrate electron density (ρe) and effective atomic numbers (Zeff) for accurate ρ determination.
• To eliminate the need for explicit tissue segmentation in DECT-based ρ imaging.

Main Methods:

• The DEEDZ-MD method establishes an empirical quadratic relationship between the atomic number-to-mass ratio and effective atomic number (Zeff) across human tissues.
• Numerical evaluations were performed using reference human tissues from ICRP Publication 110 and ICRU Report 46, with attenuation coefficients from the XCOM database.
• The method was validated using experimental DECT data from tissue characterization and anthropomorphic phantoms.

Main Results:

• A universal quadratic relation was identified for the atomic number-to-mass ratio and Zeff in human tissues.
• Simulated ρ values showed excellent agreement with reference values (0.260–3.225 g/cm³), with relative deviations within ±0.6% for most tissues.
• Experimental DECT data confirmed the method's accuracy, demonstrating practical feasibility and improved image quality (reduced noise and beam-hardening artifacts).

Conclusions:

• The DEEDZ-MD method effectively generates mass density (ρ) images from DECT data.
• This method bypasses the need for complex tissue segmentation, simplifying quantitative DECT analysis.
• The DEEDZ-MD method offers a practical and accurate approach for deriving ρ from DECT, enhancing diagnostic capabilities.