2026-09-18T17:10:22Zhttps://scholarsbank.uoregon.edu/server/oai/request

oai:scholarsbank.uoregon.edu:1794/324652026-04-17T07:00:34Zcom_1794_7557com_1794_7555com_1794_7552col_1794_169

Guldberg, Bob Hajarizadeh, Auveen 2026-04-16T15:11:11Z 2026 https://hdl.handle.net/1794/32465 0009-0008-3563-7980 87 pages. Fracture healing is a complex, multiscale process governed by tightly coordinated interactions between mechanical stimuli and biological responses. Despite bone’s intrinsic regenerative capacity, approximately 5-10% of fractures progress to nonunion, often resulting in prolonged disability, repeated surgical intervention, and substantial healthcare costs. Current clinical approaches rely primarily on radiographic assessment and nonspecific inflammatory markers, which lack sensitivity for early detection of impaired healing or infection. This diagnostic delay limits opportunities for timely intervention and highlights the need for prognostic tools capable of identifying healing trajectories at earlier stages. The objective of this thesis was to leverage predictive modeling and systemic immune profiling to identify mechanistic determinants of healing outcomes and develop translational frameworks for prognostic assessment in orthopaedic trauma. This work integrates preclinical and clinical investigations across three interconnected aims. First, rehabilitation parameters were systematically evaluated in a rat segmental bone defect model to determine how mechanical loading profiles influence regeneration. Continuous activity monitoring combined with nonlinear predictive modeling revealed that healing outcomes depend not solely on cumulative mechanical load, but on the temporal structure of loading, with rest emerging as a critical determinant. Genetic programming identified nonlinear relationships between activity parameters and bone formation, demonstrating the existence of an optimal rehabilitation zone characterized by moderate loading and sufficient recovery. Second, longitudinal immune profiling was performed using flow cytometry to characterize systemic immune dynamics following injury and rehabilitation. Distinct temporal patterns in circulating immune cell populations—including monocytes, myeloid-derived suppressor cells, and T cell subsets—were observed across healing trajectories. Rehabilitation protocols incorporating structured rest were associated with immune profiles indicative of more favorable regenerative outcomes, suggesting that mechanical loading modulates systemic immune responses that may influence healing success. Finally, these mechanistic insights were translated to a clinical population of orthopaedic trauma patients. Peripheral blood samples were analyzed using high-dimensional spectral flow cytometry to quantify immune cell populations longitudinally. Multivariate modeling using sparse partial least squares discriminant analysis identified coordinated immune signatures capable of distinguishing patients who developed fracture-related infection from those who healed successfully. These findings demonstrate the feasibility of using systemic immune profiling and predictive modeling to identify prognostic biomarkers of healing outcomes. Collectively, this thesis establishes a translational framework integrating rehabilitation mechanics, immune biology, and multivariate modeling to improve prognostic assessment in orthopaedic trauma. By identifying immune signatures and mechanical parameters associated with healing outcomes, this work provides foundational evidence supporting the development of personalized rehabilitation strategies and early diagnostic tools aimed at improving patient recovery and reducing complications. en_US University of Oregon CC BY-NC-ND 4.0 Orthopaedic Trauma Immunology Rehabilitation Bioengineering Integrating Rehabilitation Mechanics and Immune Profiling to Inform Outcomes in Orthopaedic Trauma Dissertation or thesisopen.access