Compression of CT sinogram data by decimation in the view direction.

Area of Science:

• Medical Imaging
• Image Processing
• Computed Tomography

Background:

• Clinical computed tomography (CT) acquires image data during continuous gantry rotation.
• Long integration times (frame times) can cause azimuthal blur in CT images.
• High angular sampling rates can mitigate blur but increase data volume, posing challenges for limited bandwidth systems.

Purpose of the Study:

• To evaluate the benefit of maintaining a high angular sampling rate on the CT gantry.
• To assess the efficacy of decimation (digital low-pass filtration and downsampling) in the view direction for data compression before transfer bottlenecks.
• To explore decimation as an alternative to increasing frame times in CT systems with limited bandwidth.

Main Methods:

• Theoretical evaluation of decimation effects on CT data.
• Discussion of digital filter implementation for decimation.
• Evaluation of the compression scheme using CATPHAN® 504 and human skull phantom image data.

Main Results:

• Digital decimation effectively compresses CT data before read-out while maintaining high data fidelity.
• The method preserves details in reconstructions of resolution patterns and phantom data.
• Decimation successfully prevents aliasing artifacts, offering an alternative to longer frame times.

Conclusions:

• Decimation in the view direction is a viable alternative to increasing frame times for CT systems with limited data read-out bandwidth.
• This technique can either remove aliasing artifacts or preserve spatial resolution.
• Implementing this compression scheme on the CT gantry reduces data transfer bandwidth requirements.