Improved Optical Tissue Model for Tissue Oximetry Imaging Applications

Abstract

Introduction: Chronic, non-healing wounds are a growing concern in healthcare delivery. Tissue oxygenation is recognised as critical to successful wound healing. However, the quality and quantity of the information extracted by hyperspectral imaging depend on the optical tissue model. This article aims to develop a simplified and computationally efficient approach comparable in quality with the two-layer model.

Methods: We have considered the epidermal layer as a 'thin-film' within the dermal layer. By considering the mismatched boundary and developing a four-flux model for light transport within the tissue, we have obtained a quasi two-layer model with a closed-form solution similar to the single-layer model.

Results: We have compared the developed model with the two-layer model (reference) and the single-layer model for the broad range of physiologically relevant parameters. The thickness of the epithelium: 50, 80, and 120 μm. Melanin concentration: 1, 2, 4, 8, 16, and 32%. Blood concentration: 0.2%, 1%, and 7%. Oxygen saturation: 60%, 80%, and 99%. Our initial results show that the accuracy of the proposed quasi two-layer model significantly (by a factor of 10) outperforms the single-layer model and is in close agreement with the two-layer model.

Conclusions: The proposed quasi two-layer model significantly (by the factor of 10) outperforms the single-layer model and is closely aligned with the two-layer model.

Keywords: Hyperspectral imaging; Quasi two-layer model; Skin layers; Tissue optics; Turbid tissues.