Developing a fiber-based diffuse reflectance spectroscopy setup for tissue optical property estimation

Authors:
Publisher website
Google Scholar
Abstract
Our study demonstrates that a compact, low-cost, phantom-calibrated diffuse reflectance spectroscopy system can provide realistic estimates of tissue optical properties, supporting its progression toward in situ cartilage diagnostics. We developed and validated a fiber-based DRS system capable of quantitative estimation of absorption coefficient (µ a ) and reduced scattering coefficient (µ s ′) in cartilage using a phantom-derived calibration model at wavelengths of 660, 780, and 850 nm. Diffuse reflectance spectra were acquired using a custom-built multi-distance probe, corrected using wavelength-specific calibration coefficients derived from optical phantoms, and fitted using diffusion theory with extrapolated boundary correction to recover µ a and µ s ′. Our system was validated using phantoms with known optical properties, calibrated to correct system response, and applied to articular cartilage. Theoretical and measured reflectance showed an excellent agreement (bias ∼0, 95% limits of agreement ±0.0114). After calibration, µ a and µ s ′ were extracted from bovine patellar cartilage at the same wavelengths. The estimated µ a values (0.07-0.24 cm −1 ) were characteristic of weakly absorbing, hydrated soft tissue, while µ s ′ values (9.7-15.8 cm −1 ) showed the expected consistent decrease with wavelength. We found our estimated optical properties were consistent with literature trends, especially µ s ′ values, which were between the range reported for µ s ′ values from integrating-sphere and Monte Carlo-based analysis for bovine cartilage. Phantom-calibrated DRS enables accurate, reproducible estimation of cartilage optical properties, providing a validated framework for future translation toward arthroscopic optical assessment of joint health.
Funding Information
  • Jane ja Aatos Erkon Säätiö (190001)
  • Kuopio University Hospital (5041800)
  • Orion Corporation (2025)

This publication has 25 references indexed in Scilit: