Related Experiment Video
Updated: Jun 16, 2026

Addressing Practical Issues in Atomic Force Microscopy-Based Micro-Indentation on Human Articular Cartilage Explants
Published on: October 28, 2022
Degeneration dependent changes in human knee cartilage mechanical properties revealed by synchrotron tomography based
Viktor Jönsson1, Lorenzo Grassi1, Anna Gustafsson1
1Department of Biomedical Engineering, Lund University, Lund, Sweden.
Objective:
Knee osteoarthritis (OA) is a chronic joint disease associated with pain and reduced function. The mechanisms underlying OA and how mechanical properties change with the disease are not fully understood, partly due to large variability during mechanical testing of cartilage. We characterized the mechanical properties of human femoral cartilage with varying levels of degeneration using a tissue constitution specific fibril-reinforced poroelastic (FRPE) material model.
Method:
We created sample-specific finite element (FE) models based on synchrotron-based x-ray tomography to reduce the variability caused by sample geometry. For comparison, idealized FE models were also created without tomography data. Cartilage samples (n = 15) were mechanically tested in compressive stress relaxation (two steps of 15% strain) with in-situ x-ray tomography. Adjacent tissue samples were histopathologically graded (OARSI). The FRPE parameters were optimized to minimize differences between experimental and simulated stress relaxation forces. Identified material parameters were analyzed with linear regression, with OARSI grade as independent variable.
Results:
We identified reductions in both collagen (-2x) and non-fibrillar matrix stiffness (-3x) accompanied by increased permeability (+5.5x) when comparing tissue with OARSI grade 1 and 5 in tomography-based FE models. At OARSI = 1, in idealized and tomography-based models, the collagen stiffness, non-fibrillar matrix stiffness and permeability differed 20, 110 and 55% respectively. Only collagen and non-fibrillar matrix stiffness were associated with OARSI grade in idealized models.
Conclusion:
Segmentation-based models were better at detecting degeneration-related mechanical changes than idealized models. The identified parameter changes match known OA tissue developments and can be used in predictive FE knee joint models.

