Region of Interest (ROI) Computed Tomography

Area of Science:

• Medical Imaging
• Radiology
• Computed Tomography

Background:

• Current high-resolution CT (computed tomography) methods involve trade-offs between radiation dose and image artifacts.
• Full field of view (FOV) exposure leads to high radiation doses, while region of interest (ROI) exposure can cause artifacts.

Purpose of the Study:

• To develop a CT method for high-resolution 3D reconstruction of an ROI with minimal artifacts and reduced radiation dose.
• To evaluate the efficacy of a non-uniform ROI beam filter in dose reduction and artifact mitigation.

Main Methods:

• A non-uniform ROI beam filter was developed to reduce dose outside the ROI while maintaining a higher dose within the ROI.
• High-dose full-field projections of a phantom were acquired.
• Low-dose data outside the ROI were simulated by adding noise to projection data at 1/16 and 1/256 of the original dose.
• Artifacts in reconstructed ROIs were analyzed for different ROI sizes (30% and 10% FOV) and dose reduction levels.

Main Results:

• Minimal artifacts were observed in reconstructed ROIs (30% FOV) even with significant dose reduction (1/16 and 1/256).
• For smaller ROIs (10% FOV), minimal artifacts were maintained only at the 1/16 dose reduction level.
• Artifact intensity correlated with the contrast, size, and proximity of high-contrast objects outside the ROI to the axis of rotation.

Conclusions:

• CT with an ROI filter enables high-resolution reconstruction of specific regions.
• This method achieves reduced integral radiation dose while minimizing artifacts.
• The ROI filter approach offers improved spatial resolution in targeted areas.