Assessment of Image Quality of Multi-Slice Spiral CT

Ge Wang and Michael Vannier
Departments of Biomedical Engineering and Radiology
University of Iowa
(ge-wang@uiowa.edu)

Introduction

X-ray CT is a primary imaging modality, which derives sectional and volumetric images from X-ray shadows projected from an X-ray source along various directions. Multi-slice spiral CT is a major advance, in which X-ray tube rotation, patient translation and multi-slice data acquisition are simultaneously performed. Assessment of image quality of multi-slice spiral CT is important for theoretical studies and practical applications.

Problem Description

Image quality can be described in two categories: resolution and artifacts. Image resolution has three aspects: high-contrast resolution (spatial resolution) for distinguishing adjacent objects of high-contrast, low-contrast resolution (contrast resolution) for differentiating an object from its background which is similar to the object in gray-scale, and temporal resolution for resolving time-varying structures. Image noise imposes a grainy appearance due to random fluctuations of the X-ray photon flux, and is a major factor in defining low-contrast resolution. Image artifacts are structured or patterned interference over the field of view. In this project, spatial variation of image quality in multi-slice CT will be numerically and experimentally studied.

Suggested Methods

In-plane and through-plane image resolutions are measured in terms of modulation transfer function (MTF) and section sensitivity profile (SSP). Image noise is quantified as standard deviation of water background. Due to divergence of the X-ray beam and asymmetry in the spiral CT interpolation, both resolution and noise depend on measurement location in the field of view. The project will require programming to process multi-slice spiral CT images of a high-contrast resolution phantom and a water phantom. The Analyze software will be used for image analysis and visualization.

Expected Results

The spatial variation of image quality of multi-slice CT will be quantified. The results will be plotted and animated for presentation and publication. The guidelines for clinical imaging protocol optimization will be summarized.

References

1. Wang G, Li Y: Axiomatic approach for quantification of image resolution. IEEE Signal Processing Letters 6:257-258, 1999
2. Wang G, Vannier MW: Spatial variation of section sensitivity profile in helical CT. Medical Physics 21:1491-1497, 1994
3. Wang G, Vannier MW: Computerized tomography. Encyclopedia of Electrical and Electronics Engineering, Webster JG, John Wiley & Sons, 4:8-24, 1999

CT Image of the Toshiba Phantom