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

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.

Estimation of articular cartilage properties using multivariate analysis of optical coherence tomography signal.

OBJECTIVE: The aim was to investigate the applicability of multivariate analysis of optical coherence tomography (OCT) information for determining structural integrity, composition and mechanical properties of articular cartilage. DESIGN: Equine osteochondral samples (N = 65) were imaged with OCT, and their total attenuation and backscattering coefficients (µt and µb) were measured. Subsequently, the Mankin score, optical density (OD) describing the fixed charge density, light absorbance in amide I region (Aamide), collagen orientation, permeability, fibril network modulus (Ef) and non-fibrillar matrix modulus (Em) of the samples were determined. Partial least squares (PLS) regression model was calculated to predict tissue properties from the OCT signals of the samples. RESULTS: Significant correlations between the measured and predicted mean collagen orientation (R(2) = 0.75, P < 0.0001), permeability (R(2) = 0.74, P < 0.0001), mean OD (R(2) = 0.73, P < 0.0001), Mankin scores (R(2) = 0.70, P < 0.0001), Em (R(2) = 0.50, P < 0.0001), Ef (R(2) = 0.42, P < 0.0001), and Aamide (R(2) = 0.43, P < 0.0001) were obtained. Significant correlation was also found between µb and Ef (ρ = 0.280, P = 0.03), but not between µt and any of the determined properties of articular cartilage (P > 0.05). CONCLUSION: Multivariate analysis of OCT signal provided good estimates for tissue structure, composition and mechanical properties. This technique may significantly enhance OCT evaluation of articular cartilage integrity, and could be applied, for example, in delineation of degenerated areas around cartilage injuries during arthroscopic repair surgery.

Alterations in subchondral bone plate, trabecular bone and articular cartilage properties of rabbit femoral condyles at 4 weeks after anterior cruciate ligament transection.

OBJECTIVE: To quantify early osteoarthritic-like changes in the structure and volume of subchondral bone plate and trabecular bone and properties of articular cartilage in a rabbit model of osteoarthritis (OA) induced by anterior cruciate ligament transection (ACLT). METHODS: Left knee joints from eight skeletally mature New Zealand white rabbits underwent ACLT surgery, while the contralateral (CTRL) right knee joints were left unoperated. Femoral condyles were harvested 4 weeks after ACLT. Micro-computed tomography imaging was applied to evaluate the structural properties of subchondral bone plate and trabecular bone. Additionally, biomechanical properties, structure and composition of articular cartilage were assessed. RESULTS: As a result of ACLT, significant thinning of the subchondral bone plate (P < 0.05) was accompanied by significantly reduced trabecular bone volume fraction and trabecular thickness in the medial femoral condyle compartment (P < 0.05), while no changes were observed in the lateral compartment. In both lateral and medial femoral condyles, the equilibrium modulus and superficial zone proteoglycan (PG) content were significantly lower in ACLT than CTRL joint cartilage (P < 0.05). Significant alterations in the collagen orientation angle extended substantially deeper into cartilage from the ACLT joints in the lateral femoral condyle relative to the medial condyle compartment (P < 0.05). CONCLUSIONS: In this model of early OA, significant changes in volume and microstructure of subchondral bone plate and trabecular bone were detected only in the femoral medial condyle, while alterations in articular cartilage properties were more severe in the lateral compartment. The former finding may be associated with reduced joint loading in the medial compartment due to ACLT, while the latter finding reflects early osteoarthritic changes in the lateral compartment.