Empirical cupping correction: a first-order raw data precorrection for cone-beam computed tomography.

Area of Science:

• Medical Imaging
• Computational Physics
• Biomedical Engineering

Background:

• Computed Tomography (CT) imaging is susceptible to cupping artifacts caused by nonlinearities in projection data.
• These artifacts, often stemming from beam hardening, can distort CT values and obscure details, particularly in biological samples.
• Accurate CT values are crucial for quantitative analysis and diagnosis.

Purpose of the Study:

• To introduce a novel Empirical Cupping Correction (ECC) algorithm for CT.
• To correct for cupping artifacts without requiring prior knowledge of X-ray spectrum or attenuation coefficients.
• To improve the accuracy of CT values and reduce artifacts in projection data.

Main Methods:

• The ECC algorithm is a raw data-based, empirical method that linearizes attenuation data using a polynomial precorrection function.
• Polynomial coefficients are determined once using a calibration scan of a homogeneous phantom and computed in the image domain.
• The method involves fitting basis images to a template image derived from the uncorrected phantom, without assumptions on phantom size or position.

Main Results:

• The ECC algorithm successfully precorrects raw CT data, leading to a linearization of attenuation values.
• Demonstrated application of ECC for water precorrection in an in vivo micro-CT scanner.
• Showed a significant reduction in bone-induced artifacts in in vivo mouse images, despite being a first-order correction.

Conclusions:

• ECC offers an effective, data-driven approach to mitigate CT cupping artifacts.
• The algorithm provides well-calibrated CT values and reduces image artifacts without complex spectral information.
• ECC can be combined with analytical techniques for a hybrid approach, enabling channel-dependent corrections.