Machine learning augmented near-infrared spectroscopy: In vivo follow-up of cartilage defects.

ABSTRACT

OBJECTIVE: To assess the potential of near-infrared spectroscopy (NIRS) for in vivo arthroscopic monitoring of cartilage defects. METHOD: Sharp and blunt cartilage grooves were induced in the radiocarpal and intercarpal joints of Shetland ponies and monitored at baseline (0 weeks) and at three follow-up timepoints (11, 23, and 39 weeks) by measuring near-infrared spectra in vivo at and around the grooves. The animals were sacrificed after 39 weeks and the joints were harvested. Spectra were reacquired ex vivo to ensure reliability of in vivo measurements and for reference analyses. Additionally, cartilage thickness and instantaneous modulus were determined via computed tomography and mechanical testing, respectively. The relationship between the ex vivo spectra and cartilage reference properties was determined using convolutional neural network. RESULTS: In an independent test set, the trained networks yielded significant correlations for cartilage thickness (ρ = 0.473) and instantaneous modulus (ρ = 0.498). These networks were used to predict the reference properties at baseline and at follow-up time points. In the radiocarpal joint, cartilage thickness increased significantly with both groove types after baseline and remained swollen. Additionally, at 39 weeks, a significant difference was observed in cartilage thickness between controls and sharp grooves. For the instantaneous modulus, a significant decrease was observed with both groove types in the radiocarpal joint from baseline to 23 and 39 weeks. CONCLUSION: NIRS combined with machine learning enabled determination of cartilage properties in vivo, thereby providing longitudinal evaluation of post-intervention injury development. Additionally, radiocarpal joints were found more vulnerable to cartilage degeneration after damage than intercarpal joints.