Prediction of Articular Cartilage Markers With Multispectral Imaging

Area of Science:

• Biomedical optics and imaging
• Osteoarthritis diagnostics
• Articular cartilage characterization

Background:

• Osteoarthritis (OA) is a degenerative joint disease.
• Early microstructural and compositional changes in articular cartilage (AC) are difficult to detect with current methods.
• Need for advanced diagnostic tools for early OA detection.

Purpose of the Study:

• To evaluate multispectral imaging (MSI) as a tool for characterizing articular cartilage (AC).
• To test if MSI can predict key AC properties like thickness, proteoglycan (PG) content, collagen orientation, and cell morphology.
• To assess MSI's potential for early detection of degenerative changes in AC.

Main Methods:

• Developed and evaluated a custom-built MSI system operating at six wavelengths (550-970 nm).
• Acquired reflectance images of bovine patellar AC.
• Trained machine learning models to predict AC thickness, PG content, collagen fiber orientation, and cell morphological properties (area, circularity).

Main Results:

• MSI-based machine learning models provided reliable estimates for AC thickness.
• Moderate accuracy was achieved in predicting PG content, collagen fiber orientation, and cell circularity.
• MSI demonstrated potential as a label-free method for AC characterization.

Conclusions:

• MSI holds promise as a non-invasive, label-free technique for characterizing articular cartilage.
• The developed MSI approach can aid in detecting early degenerative changes associated with osteoarthritis.
• Further research can refine MSI models for more accurate prediction of AC compositional and microstructural properties.