Assessment of cartilage thickness utilising reflectance spectroscopy

Abstract

A new principle for cartilage layer thickness assessments in joints is presented. It is based on the differences between the absorption spectra of cartilage and subchondral bone (containing blood). High-resolution ultrasound measurements of cartilage thickness were compared with reflection spectroscopy data from the same area of bovine hip joint condyles. A simple mathematical model allowed calculation of thickness and comparison with ultrasound data. The cartilage thickness was changed by being ground in short episodes. For thicker cartilage layers, a high degree of reflection in the 400-600 nm wavelength interval was seen. For thinner cartilage layers, the characteristics of the spectra of blood and bone dominated those of cartilage. The mean (+/- SD) thickness of intact cartilage was 1.21 +/- 0.30 mm (n = 30). In an exponential regression model, spectroscopic estimation of cartilage thickness showed a correlation coefficient of r = 0.69 (n = 182). For thinner cartilage layers (d < 0.5 mm), the mean model error was 0.19 +/- 0.17 mm. Results from a bi-layer Monte Carlo simulation supported the assumption of an exponential relationship between spectroscopy data and reference ultrasound data. The conclusion is that optical reflection spectroscopy can be used for cartilage layer thickness assessment.