Spectral µCT with an energy resolving and interpolating pixel detector.

Area of Science:

• Medical imaging
• Materials science
• Physics

Background:

• Laboratory x-ray micro-computed tomography (µCT) faces challenges with broad spectral content, causing reconstruction artifacts and hindering quantitative analysis.
• Monochromatization is impractical for low-brightness sources, making broad energy bandpass recording essential.

Purpose of the Study:

• To demonstrate the suitability of a new generation of pixel detectors for spectral CT.
• To show how spectral CT can overcome limitations of conventional CT by discriminating data into spectral bins.
• To validate the use of spectral CT for automated material classification and improved quantitative reconstruction.

Main Methods:

• Implementation of spectral CT using charge integrating and interpolating pixel detectors with 10 µm resolution.
• Acquisition of spectral data across multiple photon energy bins.
• Automated material classification based on spectral information.
• Reconstruction parameter tailoring for specific energy bins to account for phase contrast effects.

Main Results:

• The new pixel detectors are well-suited for implementing spectral CT at high resolution.
• Information from multiple photon energy bins significantly aids automated material classification (bones, soft tissues, metal implants).
• Spectral CT allows for better accounting of phase contrast effects, enhancing reconstruction quality.

Conclusions:

• Spectral CT with advanced pixel detectors overcomes major challenges in laboratory x-ray µCT.
• This technique facilitates automated material discrimination and improves quantitative reconstruction accuracy.
• The ability to tailor reconstruction to energy bins enhances phase contrast analysis, leading to more reliable imaging results.