Quantitative principal component model for skin chromophore mapping using multi-spectral images and spatial priors

Area of Science:

• Biomedical Optics
• Medical Imaging Analysis
• Quantitative Photo-physics

Background:

• Accurate quantification of blood volume and oxygenation is crucial for diagnosing and monitoring various skin conditions.
• Existing multi-spectral imaging analysis methods often lack quantitative accuracy or require invasive procedures.
• Principal Component Analysis (PCA) offers a potential avenue for robust data analysis in complex optical imaging.

Purpose of the Study:

• To develop and validate a novel Principal Component Analysis (PCA)-based reconstruction algorithm for quantitative blood volume and oxygenation extraction from multi-spectral imaging data.
• To assess the accuracy and limitations of the PCA method, particularly concerning the influence of epidermal thickness.
• To demonstrate the in vivo applicability of the PCA algorithm for skin analysis.

Main Methods:

• Application of PCA to multi-spectral imaging data from numerical phantoms based on a two-layered skin model.
• Derivation of analytical expressions to convert PCA results into quantitative blood volume and oxygenation values.
• Evaluation of the method's accuracy with known and unknown epidermal thickness variations.
• In vivo validation using multi-spectral and Optical Coherence Tomography (OCT) imaging of a healthy volunteer's forearm.

Main Results:

• The PCA-based algorithm accurately extracts blood volume with less than 6% error, even with a 0.04mm deviation in epidermal thickness.
• Blood oxygenation extraction shows higher sensitivity to epidermal thickness variations, with errors up to 25% for the same deviation.
• When the underlying skin structure is known, the PCA reconstruction achieves less than 8% error for both blood volume and oxygenation.
• Point-wise correlation between PCA-based reconstruction and analytical model results from in vivo data confirms the method's proof of principle.

Conclusions:

• PCA-based reconstruction provides a viable and quantitative method for extracting blood volume and oxygenation from multi-spectral skin imaging.
• The algorithm demonstrates robustness for blood volume estimation, with potential for clinical application in non-invasive skin diagnostics.
• Further refinement may be needed to mitigate errors in oxygenation quantification related to epidermal thickness variability.