Registration scheme suitable to Mueller matrix imaging for biomedical applications.

Area of Science:

• Biomedical optics
• Image processing
• Polarimetry

Background:

• Mueller matrix imaging polarimeters typically require sequential acquisition of multiple raw images.
• In vivo applications are challenged by subject motion, causing image shifts and distortions that prevent accurate Mueller image extraction.
• Existing methods struggle with spatially inhomogeneous and depolarizing samples common in biological tissues.

Purpose of the Study:

• To develop and validate a registration algorithm for correcting motion artifacts in in vivo Mueller matrix imaging.
• To enable reliable extraction of Mueller images despite subject movement and sample depolarization.
• To address limitations of current techniques in dynamic biological imaging.

Main Methods:

• Implementation of a registration algorithm based on optical flow principles.
• Sequential acquisition of raw images with varying polarization states.
• Experimental validation using pig skin samples under static and dynamic conditions.

Main Results:

• The optical flow-based registration algorithm effectively corrects motion-induced distortions in raw images.
• Accurate Mueller images were successfully extracted from dynamic acquisitions of biological tissue.
• The algorithm demonstrates robustness in the presence of medium to strong depolarization.

Conclusions:

• The developed optical flow algorithm significantly improves the feasibility of in vivo Mueller matrix imaging.
• This technique overcomes major challenges posed by subject motion and sample depolarization.
• It paves the way for advanced polarimetric studies of dynamic biological processes.