Boston University
A Raman-based classification framework for anatomically resolved mapping of articular cartilage health
Abstract
dc:description.abstractPost-traumatic osteoarthritis (PTOA) is characterized by early biochemical alterations in the extracellular matrix (ECM) of articular cartilage, including glycosaminoglycan (GAG) depletion, collagen disruption, and changes in tissue hydration, which precede visible morphological degeneration. Current clinical diagnostics primarily assess structural damage and do not provide quantitative evaluation of cartilage composition or function, limiting early detection and treatment selection. This study presents a Raman-based classification framework for anatomically resolved assessment of cartilage composition and mechanical function.Osteochondral samples were obtained from human knee joints (n = 17 donors) and systematically mapped across anatomically distinct regions, including femoral condyles (anterior, central, posterior) and tibial plateaus (meniscus-covered and uncovered regions), on both medial and lateral compartments. A handheld, arthroscopy-compatible Raman probe (785 nm excitation) was used to acquire spatially resolved spectra across joint surfaces. Spectra were preprocessed using polynomial baseline correction and normalized to the fingerprint region (800–1800 cm⁻¹), followed by multivariate linear decomposition to extract relative ECM biomarker scores corresponding to GAG, collagen, and water content. Mechanical properties were quantified via spherical indentation testing (3 mm diameter indenter; 50–200 µm displacement; 500 s relaxation), and elastic modulus was calculated using a Hertzian contact model. Raman-derived biomarker combinations were correlated with measured mechanical properties, and nonlinear regression models were developed to predict cartilage stiffness from compositional features. These models enabled generation of spatial maps of predicted mechanical properties across anatomically defined regions of each joint. Significant regional variation in both composition and mechanics was observed, with trends including reduced GAG content and mechanical stiffness in posterior femoral regions and increased variability in tibial uncovered regions. Raman-derived biomarkers demonstrated strong correlations with mechanical properties (R² ≈ 0.6–0.7), supporting their ability to capture functionally relevant tissue changes. A region-specific quartile-based classification framework was implemented by partitioning predicted mechanical values within each anatomical region into four equal groups, enabling relative stratification of cartilage health. This approach revealed distinct spatial patterns of tissue integrity within and across joints, differentiating regions of preserved versus compromised cartilage not evident through gross inspection alone. This work establishes a quantitative, anatomically resolved framework linking ECM composition to functional mechanical behavior, enabling spatial classification of cartilage health. The integration of Raman spectroscopy with a fiber-optic probe platform supports its potential for intraoperative application, providing real-time, data-driven assessment of cartilage quality to improve clinical decision-making and guide personalized treatment strategies.
Degree
thesis:*- Name thesis:degree_name
- Master of Science
- Level thesis:degree_level
- masters
- Discipline thesis:degree_discipline
- Biomedical Engineering
- Grantor
- Boston University
- Year dc:date.issued
- 2026
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Garcia Estrada, Javier Armando
- Advisor dc:contributor.advisor
-
- Albro, Michael
- ORCID dc:identifier.orcid
Subjects
dc:subject × 3Rights
- Language dc:language.iso
- en_US
Identifiers
dc:identifier.*- Handle dc:identifier.uri
- https://hdl.handle.net/2144/52980