Effect of collagen fibril distributions on the crack profile in articular cartilage

Integration of Swin UNETR and statistical shape modeling for a semi-automated segmentation of the knee and biomechanical modeling of articular cartilage

Simulation studies, such as finite element (FE) modeling, provide insight into knee joint mechanics without patient involvement. Generic FE models mimic the biomechanical behavior of the tissue, but overlook variations in geometry, loading, and material properties of a population. Conversely, subject-specific models include these factors, resulting in enhanced predictive precision, but are laborious and time intensive. The present study aimed to enhance subject-specific knee joint FE modeling by incorporating a semi-automated segmentation algorithm using a 3D Swin UNETR for an initial segmentation of the femur and tibia, followed by a statistical shape model (SSM) adjustment to improve surface roughness and continuity. For comparison, a manual FE model was developed through manual segmentation (i.e., the de-facto standard approach). Both FE models were subjected to gait loading and the predicted mechanical response was compared. The semi-automated segmentation achieved a Dice similarity coefficient (DSC) of over 98% for both the femur and tibia. Hausdorff distance (mm) between the semi-automated and manual segmentation was 1.4 mm. The mechanical results (max principal stress and strain, fluid pressure, fibril strain, and contact area) showed no significant differences between the manual and semi-automated FE models, indicating the effectiveness of the proposed semi-automated segmentation in creating accurate knee joint FE models. We have made our semi-automated models publicly accessible to support and facilitate biomechanical modeling and medical image segmentation efforts ( https://data.mendeley.com/datasets/k5hdc9cz7w/1 ).

A numerical model for fibril remodeling in articular cartilage

BACKGROUND

Collagen fibrils of articular cartilage have a distinct organization in mature human knee joints. It seems that a mechanobiological process drives the remodeling of newborn collagen fibrils with maturation. Therefore, the goal of the present study was to develop a collagen fibril remodeling algorithm that describes the unique collagen fibril organization in a 3D knee model.

METHOD

A fibril-reinforced, biphasic cartilage model was used with a cuboid and a 3D human knee joint geometries. An isotropic collagen fibril distribution was assigned to the cartilage at the start of the analysis. Each fibril was rotated towards the direction that resulted in a maximum stretch at each time increment of the loading cycle.

RESULTS

The resulting pattern for the collagen fibrils was compared with split line patterns of porcine knee joint cartilage and also data published in the literature. Fibrils on the articular surface had a radial pattern towards the geometrical centroid of the tibial and femoral cartilage. In the tibiofemoral contact regions of superficial zone, fibrils were oriented circumferentially and randomly. In the porcine samples, the split-line patterns were similar to those obtained theoretically. Depth-wise organization of fibril network was characterized by fibrils perpendicular to the subchondral bone in the deeper layers, and fibrils parallel to the surface of cartilage in the superficial zone.

CONCLUSIONS

The maximum stretch criterion, coupled with a biphasic constitutive model, successfully predicted the collagen fibril organization observed in the articular cartilage throughout the depth and on the articular surface.

The effect of highly inhomogeneous biphasic properties on mechanical behaviour of articular cartilage

BACKGROUND AND OBJECTIVE

Investigating the biomechanics of cartilage could help to understand the unique load-bearing property of the cartilage and optimize the scaffold design in tissue-engineering. It is important to model the cartilage as a highly inhomogeneous fibril-reinforced biphasic material to represent its complex composition and structure. The depth-dependent and strain-dependent properties of the cartilage would also play an important role in its mechanical behaviour. However, the differences in representing the cartilage as a highly inhomogeneous model or as simplified models still remain unclear. Hence, in this study, a highly inhomogeneous fibril-reinforced biphasic cartilage model considering both the depth-dependent and strain-dependent properties was constructed; the effect of highly inhomogeneous properties on the mechanical behaviour of articular cartilage was investigated.

METHODS

A finite element model of the cartilage was developed based on a flat-ended indentation test. Compressive forces were applied to four various inhomogeneous layered models through a porous indenter (Model 1: nine layers with strain-dependent permeability; Model 2: three layers with strain-dependent permeability; Model 3: single layer with strain-dependent permeability; Model 4: nine layers with constant permeability).

RESULTS

Models 1 and 2 provided similar results with less than 3% difference in the peak effective stress, contact pressure, fluid pressure as well as fluid support ratio. However, Model 1 to Model 3 differed in stress and strain distribution patterns along depth over prolonged loads, which may provide an important insight into the highly inhomogeneous depth-dependent properties of cartilage. In addition, Model 1 with strain-dependent permeability demonstrated an enhanced capability on fluid pressurisation as compared with Model 4 which had constant permeability.

CONCLUSIONS

A highly inhomogeneous fibril-reinforced biphasic model considering both depth-dependent and strain-dependent properties was developed in this study, in order to illustrate the effect of highly inhomogeneous properties on the mechanical behaviour of the articular cartilage. The number of layers in the models with depth-dependent properties should be selected according to the research questions and clinical demands. The model with strain-dependent permeability offers an enhanced capability on fluid pressurisation. In future studies, the proposed model could be adopted in cell-models to provide more in-depth information or in tissue-engineering to optimize the depth-dependent scaffold structure.

A Probabilistic Failure Risk Approach to The Problem of Articular Cartilage Lubrication

BACKGROUND AND OBJECTIVE

The geometrical and mechanical properties that characterise the cartilage contact gap are uncertain and spatially varied. To date the effects of such uncertainties on cartilage lubrication have not been explored. Using a probabilistic approach, the purpose of this study is to numerically investigate the influence of surficial cartilage glycoaminoglycan (GAG) content on joint lubrication behaviour. Gap asperity stiffness and polymer brush border (PBB) thickness are affected by the uncertainty of surficial GAG concentration, and so their correlated effects in maintaining hydrodynamic joint lubrication are investigated.

METHODS

Correlated sampling data are first generated by Monte Carlo simulation. These data are used as inputs for the cartilage contact model, which includes three distinctive features of cartilage tissue (tension-compression nonlinearity, aggrecan dependent permeability and compressive modulus) and fluid flow resistance effects of PBB on cartilage surface. The degree of hydrodynamic lubrication after thirty minutes of constant loading is used as an indicator for assessing the lubrication performance at the contact interface.

RESULTS

The increase of PBB thickness with GAG concentration enhances the hydrodynamic lubrication component in the cartilage contact gap, whereas increasing the asperity stiffness with GAG concentration impairs hydrodynamic lubrication. GAG loss rate increases with the rise of GAG concentration. More aggrecan shedding through the surface could result in a thicker and denser PBB, and therefore enhance the lubrication performance in mixed-mode regime. On the other hand, higher GAG content makes the asperities stiffer, which may impede contact gap closure, and thus encourage gap fluid loss and impair the lubrication performance of cartilage.

CONCLUSION

The lubrication performance of cartilage varies with the physiological conditions of the joint. Since a range of variables are internally related, the outcomes on joint lubrication are difficult to predict. A probabilistic approach accounting for the uncertainties can potentially result in more accurate evaluations of joint lubrication performance.