Assessment of hybrid speckle reduction algorithms

Area of Science:

• Medical Imaging
• Ultrasound Technology
• Signal Processing

Background:

• Coherent detection in medical ultrasound imaging causes interference, leading to speckle artefacts.
• Speckle artefacts significantly degrade the information content of ultrasound images.
• Existing speckle reduction techniques, such as spatial/frequency compounding and image filtering, have limitations.

Purpose of the Study:

• To introduce and evaluate novel speckle reduction algorithms for ultrasound pulse-echo images.
• To address the shortcomings of traditional speckle reduction methods.
• To assess the potential for real-time implementation of new speckle reduction techniques.

Main Methods:

• Development of two new speckle reduction algorithms: 'local frequency diversity' and 'frequency differencing'.
• Deterministic identification and correction of speckle within localized corrupted segments of A-lines.
• Assessment of algorithms using simulated and clinical in vivo ultrasound images.

Main Results:

• The proposed algorithms effectively identify and correct speckle artefacts.
• The localized correction approach minimizes degradation of image information content.
• The techniques show potential for real-time processing in ultrasound systems.

Conclusions:

• Local frequency diversity and frequency differencing offer an alternative to conventional speckle reduction methods.
• These novel algorithms provide targeted speckle correction, preserving image details.
• The developed techniques hold promise for enhancing real-time medical ultrasound imaging quality.