New method of evaluating edge-preserving adaptive filters for computed tomography (CT): digital phantom method

Area of Science:

• Medical Imaging
• Image Processing
• Radiology

Background:

• Low-dose scanning in medical imaging necessitates noise reduction techniques.
• Maintaining image resolution and quality is crucial for accurate diagnosis.

Purpose of the Study:

• To evaluate edge-preserving adaptive filters for noise elimination in low-dose scanning.
• To assess the impact of these filters on image resolution and noise characteristics.

Main Methods:

• Development of a digital phantom image for controlled evaluation.
• Analysis of noise statistical values, noise characteristics, and resolution characteristics.
• Assessment of filter performance in low-contrast and high-contrast regions.

Main Results:

• Edge-preserving adaptive filters act as smoothing filters in noisy, low-contrast regions, enhancing density resolution.
• Filter strength is reduced in high-contrast regions to preserve spatial resolution of small structures.
• Filters effectively target noise amplified by reduced exposure doses, improving overall image quality.

Conclusions:

• Edge-preserving adaptive filters are effective for selective noise elimination in low-dose imaging.
• Digital phantom images provide a versatile tool for evaluating image processing techniques.
• These filters significantly enhance image quality in challenging low-dose scanning conditions.