An Image-Domain Contrast Material Extraction Method for Dual-Energy Computed Tomography

Area of Science:

• Medical Imaging
• Radiology
• Image Processing

Background:

• Conventional dual-energy CT material decomposition relies on mass/volume conservation.
• This limits the accurate assignment of CT numbers to specific materials.
• A novel approach is needed for preferential material extraction.

Purpose of the Study:

• To introduce and evaluate an image-domain contrast material extraction process (CMEP).
• To demonstrate CMEP's ability to isolate contrast-producing materials in dual-energy CT.
• To assess CMEP's performance in phantom and in vivo experiments.

Main Methods:

• CMEP utilizes image processing freeware (Fiji) for arithmetic operations on dual-energy ratio maps.
• Material-specific masks are generated based on ratio intervals.
• Masks are applied to original CT images to create material-specific images.
• Quantitative analysis used phantoms with tungsten, calcium, and iodine; qualitative analysis used a rabbit model.

Main Results:

• Distinct dual-energy ratios were observed for tungsten, iodine, and calcium, largely independent of concentration.
• CMEP successfully separated materials in both phantom and in vivo studies.
• Maximum errors for pure materials were low (e.g., 0.9 mg/mL for iodine).
• Mixtures showed some discrepancies, particularly iodine and calcium, but vasculature remained clear in rabbit models.

Conclusions:

• CMEP offers a straightforward, image-domain method for material signal extraction in dual-energy CT.
• It is valuable for separating high-Z contrast elements from conventional contrast agents like iodine or calcium.
• The method is implementable in freeware and adaptable across vendors.