A novel automatic 3D stitching algorithm for optical coherence tomography angiography and its application in...

Area of Science:

• Biomedical Imaging
• Dermatology
• Optical Engineering

Background:

• Optical Coherence Tomography (OCT) and OCT Angiography (OCTA) are valuable for skin imaging.
• Current OCT/OCTA systems have limited fields of view (FOV), hindering comprehensive skin lesion assessment.
• Expanding FOV with wide-field lenses often compromises lateral resolution.

Purpose of the Study:

• To develop and validate an automated 3D stitching method for large FOV, high-resolution OCT/OCTA skin imaging.
• To overcome the resolution-FOV trade-off in current OCT/OCTA skin imaging techniques.
• To enable detailed visualization of skin structures and vasculature over larger areas.

Main Methods:

• Developed an automated 3D stitching algorithm to montage adjacent OCT and OCTA volumes.
• Acquired 3x3 OCT/OCTA datasets (nine volumes per dataset) from healthy skin of six volunteers.
• Evaluated image quality using structural similarity index, feature similarity index, and vessel contrast-noise-ratio (CNR).

Main Results:

• The automated stitching method successfully created high-resolution, large FOV skin volumes.
• Stitched images demonstrated superior similarity to single-scan protocols for large-scale imaging.
• Vessel CNR improved significantly from 2.07 ± 0.44 (single scan) to 3.05 ± 0.51 (stitched method).

Conclusions:

• The proposed 3D stitching method offers a flexible and repeatable solution for large FOV OCT/OCTA skin imaging.
• This technique effectively enhances image resolution and clarity for detailed skin structure and vascular analysis.
• The method overcomes limitations of current systems, paving the way for improved dermatological diagnostics.