Infrastructure Engineering - Research Publications

Permanent URI for this collection

http://hdl.handle.net/11343/347

Filters

2020 - 2023 3
3
Reset filters

Settings
Statistics
Citations
Author: Liu, Xuanchi × Author: Miramini, Seyedsaeed × Author: Zhang, Lihai × Start date: 2000 ×

Search Results

Now showing 1 - 3 of 3
  • Item
    No Preview Available
    Development of numerical model-based machine learning algorithms for different healing stages of distal radius fracture healing
    Liu, X ; Miramini, S ; Patel, M ; Ebeling, P ; Liao, J ; Zhang, L (ELSEVIER IRELAND LTD, 2023-05)
    BACKGROUND AND OBJECTIVES: Early therapeutic exercises are vital for the healing of distal radius fractures (DRFs) treated with the volar locking plate. However, current development of rehabilitation plans using computational simulation is normally time-consuming and requires high computational power. Thus, there is a clear need for developing machine learning (ML) based algorithms that are easy for end-users to implement in daily clinical practice. The purpose of the present study is to develop optimal ML algorithms for designing effective DRF physiotherapy programs at different stages of healing. METHOD: First, a three-dimensional computational model for the healing of DRF was developed by integrating mechano-regulated cell differentiation, tissue formation and angiogenesis. The model is capable of predicting time-dependant healing outcomes based on different physiologically relevant loading conditions, fracture geometries, gap sizes, and healing time. After being validated using available clinical data, the developed computational model was implemented to generate a total of 3600 clinical data for training the ML models. Finally, the optimal ML algorithm for each healing stage was identified. RESULTS: The selection of the optimal ML algorithm depends on the healing stage. The results from this study show that cubic support vector machine (SVM) has the best performance in predicting the healing outcomes at the early stage of healing, while trilayered ANN outperforms other ML algorithms in the late stage of healing. The outcomes from the developed optimal ML algorithms indicate that Smith fractures with medium gap sizes could enhance the healing of DRF by inducing larger cartilaginous callus, while Colles fractures with large gap sizes may lead to delayed healing by bringing excessive fibrous tissues. CONCLUSIONS: ML represents a promising approach for developing efficient and effective patient-specific rehabilitation strategies. However, ML algorithms at different healing stages need to be carefully chosen before being implemented in clinical applications.
  • Item
    No Preview Available
    Influences of variability and uncertainty in vertical and horizontal surface roughness on articular cartilage lubrication
    Liao, J ; Liu, X ; Miramini, S ; Zhang, L (PERGAMON-ELSEVIER SCIENCE LTD, 2022-09)
    BACKGROUND AND OBJECTIVES: Cartilage surface roughness has significant implications on joint lubrication. However, the effects of the variability in surface roughness in different directions (especially in horizontal direction) in mixed-mode lubrication have not been fully investigated and relevant research work in this field is limited. This study presents a probabilistic numerical approach to investigate the influence of variability and uncertainty of Root-Mean-Square (RMS) roughness heights (vertical roughness) and roughness correlation lengths (horizontal roughness) on cartilage lubrication. METHODS: The synthetic surface topographies with typical ranges of vertical and horizontal roughness characteristics were firstly input to a coupled cartilage contact model. A response surface was then constructed using the input roughness parameters and the output coefficient of friction (CoF). Finally, a large number of independent or correlated roughness samples were generated for computing the probability of mixed-mode lubrication failure (PoF), which was defined as CoF > 0.27 (corresponding to a 90% loss of fluid support in the contact interface). RESULTS: Both independent RMS roughness heights and correlation lengths are correlated positively with CoF. This indicates that the increase of the vertical surface roughness could exacerbate cartilage wear, whereas increasing surface roughness in horizontal direction (i.e., reducing correlation lengths) could retain gap fluid that aids mixed-mode lubrication. Importantly, it shows that CoF is dominant by RMS roughness height. The uncertainty in the independent correlation lengths may lead to the underestimation of PoF. By simulating osteoarthritic surface roughness with a strong correlation between RMS roughness heights and correlation lengths, the value of PoF could reach 70-99%. CONCLUSION: This study highlights the significance of incorporating the mutual relations between the surface roughness in vertical and horizontal directions into research, and the findings could potentially contribute to the design of biomimetic cartilage surfaces for the treatment of osteoarthritis.
  • Item
    Thumbnail Image
    Computational study on synovial fluid flow behaviour in cartilage contact gap under osteoarthritic condition
    Liao, J ; Miramini, S ; Liu, X ; Zhang, L (Elsevier, 2020-08-01)
    This study numerically investigates the pathological changes of fluid flow in cartilage contact gap due to the changes in cartilage surface roughness and synovial fluid characteristics in osteoarthritic (OA) condition. First, cartilage surface topographies in both healthy and OA conditions are constructed using a numerical approach with consideration of both vertical and horizontal roughness. Then, constitutive equations for synovial fluid viscosity are obtained through calibration against previous experimental data. Finally, the roughness and synovial fluid information are input into the gap flow model to predict the gap permeability. The results show that the rougher surface of OA cartilage tends to decrease gap permeability by around 30%–60%. More importantly, with the reduction in gap size, the decrease in gap permeability becomes more significant, which could result in an early fluid ultrafiltration into the tissue. Moreover, it is demonstrated that the pathological synovial fluid has more deleterious effects on the gap permeability than the OA cartilage surface, as it could potentially increase the gap permeability by a few hundred times for pressure gradients less than 106 Pa/m, which could inhibit the fluid ultrafiltration into the tissue. The outcomes from this research indicate that the change in fluid flow behaviour in contact gap in OA condition could significantly affect the function of articular joints.