Volumetric Visualization of the Dermal Vasculature with Signal and Image-based Feature Extraction on a High-frequency...

Area of Science:

• Medical Imaging
• Biomedical Engineering
• Ultrasound Technology

Background:

• 3-D volumetric imaging of skin vasculature is crucial for diagnostics.
• Continuous scanning with 1-D ultrasound arrays offers faster data acquisition but introduces noise, hindering image quality.
• Existing methods struggle with filtering flow signals and maintaining image contrast due to transducer translation noise.

Purpose of the Study:

• To develop and validate a processing algorithm for enhancing 3-D visualization of dermal vasculature.
• To overcome noise artifacts introduced by continuous scanning in high-frequency ultrasound.
• To improve the contrast-to-noise ratio for clearer imaging of skin vascular networks.

Main Methods:

• Volumetric ultrasound data acquired from the dorsum of the hand using continuous scanning.
• Sectioning of data and application of region-based singular value decomposition (SVD) filtering for flow signal extraction.
• Two-step de-noising using non-local means (NLM) and top-hat (TH) morphological filters to remove continuous scanning noise.
• Reconstruction and 3-D rendering of processed data for volumetric visualization.

Main Results:

• Section-wise SVD processing successfully highlighted vasculature in 2-D slices.
• NLM and TH filters effectively removed background noise, significantly improving image quality.
• Contrast-to-noise ratio improved from 12.87 ± 1.3 dB to 18.91 ± 2.3 dB (after NLM) and 62.61 ± 19.2 dB (after NLM + TH).

Conclusions:

• The proposed algorithm enables comprehensive 3-D visualization of dermal vasculature using a 1-D ultrasound array.
• This approach demonstrates feasibility and significance for imaging complex vascular structures in 3-D.
• The enhanced visualization techniques can improve diagnostic capabilities in skin imaging.