Calibration Free Beam Hardening Correction for Cardiac CT Perfusion Imaging

Area of Science:

• Medical Imaging
• Radiology
• Image Processing

Background:

• Computed tomography (CT) is a promising noninvasive tool for coronary angiography, but beam hardening artifacts (BHA) hinder accurate myocardial perfusion imaging using CT (MPI-CT) blood flow calculations.
• Current BHA correction (BHC) methods often require specialized energy-sensitive CT scanners or calibration procedures, limiting their widespread applicability.

Purpose of the Study:

• To develop and evaluate a calibration-free, automatic BHC (ABHC) algorithm applicable to any BHC method for improving MPI-CT.
• To assess the efficacy of the ABHC algorithm in reducing BHA in simulated, physical phantom, and preclinical MPI-CT datasets.

Main Methods:

• Developed an iterative, calibration-free ABHC algorithm utilizing a BHA-specific cost function.
• Segmented CT images into bone, iodine, and soft tissue components for targeted correction.
• Applied a polynomial BHC method extended to three materials, iteratively refining corrections based on forward and back projections.

Main Results:

• Significantly reduced BH streak artifacts in simulated phantoms (12±2HU to 1±1HU) and physical phantoms (48±6HU to 1±5HU).
• Achieved substantial reduction in cupping artifacts by 81% in simulated and 86% in physical phantoms.
• Reduced BHA in preclinical MPI-CT images from 28±6 HU to below 4±4HU at peak enhancement.

Conclusions:

• The developed ABHC algorithm effectively reduces beam hardening artifacts in CT imaging without requiring calibration or specialized scanners.
• This method shows significant potential for enhancing the accuracy of MPI-CT and improving its role as a noninvasive gatekeeper for coronary angiography.