Improving automatic contrast agent extraction system using monochromatic CT number.

Abstract

In a previous study, a phantom study of a contrast agent extraction system with computed tomography (CT) number and raw-data-based electron density (ED) was described. The current study improved this system with monochromatic CT (mCT) number and evaluated an anthropomorphic phantom for delineation of the contrast-enhanced region. Dual-energy CT images were scanned with a tissue-equivalent phantom and an anthropomorphic phantom with an iodinated contrast agent (1-130mg/mL). The 40, 70, and 130keV mCT images were reconstructed with 80 and 135kV CT images. The contrast agent was separated from other materials using the gradient of the mCT number (GmCT) and the threshold mCT numbers. The system was analyzed using in-house software with Python. The evaluation of the accuracy for the contrast agent extraction was performed by measuring the ratio of the volume (ROV). The mCT number of the contrast agent and bone materials, liver, and muscle in the tissue-equivalent phantom was obviously greater than - 78 HU. The deviation of the mCT numbers between bone materials in tissue-equivalent phantom and the contrast agent were larger than 8 HU. The GmCT was within 4.0 in the tissue-equivalent phantom and more than 6.0 in the contrast agent. The ROV was 0.97-1.00 at more than 1mg/mL contrast agent. Improved the contrast agent extraction system could be used for a patient's CT image. It could extract the iodinated tumor or lesion automatically. The contrast agent extraction system was improved by the mCT number. It is expected to only extract the contrast-enhanced region automatically.