Toward Tailored Rehabilitation by Implementation of a Novel Musculoskeletal Finite Element Analysis Pipeline

Abstract

Tissue-level mechanics (e.g., stress and strain) are important factors governing tissue remodeling and development of knee osteoarthritis (KOA), and hence, the success of physical rehabilitation. To date, no clinically feasible analysis toolbox has been introduced and used to inform clinical decision making with subject-specific indepth joint mechanics of different activities. Herein, we utilized a rapid state-of-the-art electromyographyassisted musculoskeletal finite element analysis toolbox with fibril-reinforced poro(visco)elastic cartilages and menisci to investigate knee mechanics in different activities. Tissue mechanical responses, believed to govern collagen damage, cell death, and fixed charge density loss of proteoglycans, were characterized within 15 patients with KOA while various daily activities and rehabilitation exercises were performed. Results showed more interparticipant variation in joint mechanics during rehabilitation exercises compared to daily activities. Accordingly, the devised workflow may be used for designing subject-specific rehabilitation protocols. Further, results showed the potential to tailor rehabilitation exercises, or assess capacity for daily activity modifications, to optimally load knee tissue, especially when mechanically-induced cartilage degeneration and adaptation are of interest.

An Automated and Robust Tool for Musculoskeletal and Finite Element Modeling of the Knee Joint

“…We developed the template-based MSK and FE modeling pipeline to ensure the tool consistently produced a personalized model without failure. We adopted our previously developed and validated template-based modeling [9], [17], [21], in which a template model was scaled according to the anatomical dimensions of the subject's knee, such as mediolateral and anteroposterior knee dimensions. In this current study, we used the same template MRI and FE model (with the FRPVES material model incorporated) but modified the scaling pipeline [23] to automatically transform the template MRI to the subject's MRI.…”

“…However, these tools are not fully automated [17], [21], [23], do not include whole tibiofemoral and patellofemoral joints (such as menisci) [21], [22], or do not offer MSK modeling [21]- [23]. The tool developed in this study, which has integrated MSK and FE models, can be applied to the study of knee biomechanics during different functional activities and rehabilitation exercises or knee injuries/surgeries such as ligaments and menisci [9], [31], [86]. More specifically, the FE model with FRPVES material incorporates the structure and composition of knee cartilage and menisci (such as collagen network and proteoglycan and water content) and their interplay with tissue mechanobiological response governing tissue degradation and remodeling [2], [3].…”

“…Recent research has demonstrated the potential of computational modeling, namely musculoskeletal (MSK) and finite element (FE) models, to estimate personalized knee joint mechanobiological environment and degradation response [6], [9]- [12]. Indeed, FE models have been widely used to estimate the knee's within tissue biomechanics (e.g., cartilage stress and strain) governing tissue degradation response [6]- [8], [12], [13].…”

“…The process involves image segmentation, processing of the segmented geometries, meshing, model assembly, material model incorporation, estimation and incorporation of loading and boundary conditions, and achieving a converged solution [16], [17]. This need for specialized knowledge presents a barrier to broader use and limits reproducibility [18] of MSK and FE models in An Automated and Robust Tool for Musculoskeletal and Finite Element Modeling of the Knee Joint research and clinical settings where personalized models could have a significant impact on health outcomes [9], [11], [19], [20].…”

“…Several studies have developed rapid and (semi-)automatic methods to create knee joint models. Template-based methods have enabled rapid generation of knee MSK-FE models [17], [21] with the potential to estimate KOA progression [21] or plan personalized rehabilitation [9]. However, these methods are not fully automated, e.g., they require manual measurement of knee anatomical dimensions from medical images.…”