Speckle reduction by phase-based weighted least squares.

Area of Science:

• Medical Imaging
• Signal Processing
• Biomedical Engineering

Background:

• Ultrasonography is a vital diagnostic tool.
• Speckle noise in ultrasound images significantly hinders accurate diagnosis.
• Existing noise reduction methods often degrade image features.

Purpose of the Study:

• To develop a novel framework for effective speckle reduction in ultrasound images.
• To preserve image features, such as low contrast edges, during noise removal.
• To improve diagnostic accuracy by enhancing ultrasound image quality.

Main Methods:

• A phase-based weighted least squares optimization framework was developed.
• A local phase-based measure, invariant to intensity, was used for edge map extraction.
• The extracted edge map guided the despeckling optimization process.

Main Results:

• The proposed method effectively smoothed speckle noise in synthetic and clinical ultrasound images.
• Image features, including edges with varying contrasts, were well preserved.
• The approach demonstrated superior performance compared to existing state-of-the-art methods.

Conclusions:

• The novel phase-based weighted least squares method significantly reduces speckle noise in ultrasound images.
• This technique enhances image quality by preserving crucial diagnostic features.
• The proposed approach offers a promising solution for improving ultrasound-based medical diagnoses.