Algorithm for using dual energy computed tomography to determine chemical composition: A feasibility study

Area of Science:

• Medical Imaging
• Materials Science
• Computational Physics

Background:

• Accurate material identification is crucial for various scientific and medical applications.
• Dual-energy computed tomography (CT) offers potential for material characterization.
• Existing methods may have limitations in accuracy and handling complex mixtures.

Purpose of the Study:

• To develop and validate an algorithm for identifying chemical constituents of unknown materials using dual-energy CT.
• To improve the accuracy of material discrimination based on mass attenuation coefficients (MAC).
• To assess the algorithm's performance in determining elemental composition and weight fractions.

Main Methods:

• Developed an algorithm utilizing mass attenuation coefficients (MAC) obtained from dual-energy CT scans.
• Employed dual energy settings (80/140 kVp) and mono-energetic X-rays (50/80, 80/100 keV) as inputs.
• Validated the algorithm by determining the chemical constituents of human tissues and comparing results with NIST data.

Main Results:

• The algorithm successfully identified the elemental composition and weight fractions of unknown materials.
• High accuracy was achieved, with differences below 0.01% for compounds and 6.02% for mixtures using mono-energetic X-rays.
• Spectral X-ray analysis showed differences of 2.98% for compounds and 6.03% for mixtures.

Conclusions:

• A novel algorithm effectively determines elemental type and weight fraction using dual-energy CT MAC.
• The algorithm excludes inherent uncertainty from SPR calculations, enhancing accuracy.
• This advancement promises improved dose calculations in radiation therapy planning.