Low back pain (LBP) is a common musculoskeletal complaint, and is a particularly large problem in the military. Heavy load carriage and unusual postures military members operate under in training and in combat have been implicated as primary causes of LBP. The muscles of the lumbar spine (LS) are crucial for providing stability and maintaining posture; lumbar instability is associated with chronic LBP and can result in injury. With injury and age, degenerative structural changes in muscle tissue are observed in concert with gross alterations in posture, however the relationship between muscle microstructure and posture isn’t well understood. The purpose of these studies was to investigate the predictive capacity of muscle structure on LS posture in active-duty Marines under operationally relevant conditions. Fractional anisotropy, a measure of muscle microstructure made with diffusion tensor imaging (DTI), was found to be a key predictor of LS posture. Trying to translate these findings to physiologic measures of muscle, it became evident that DTI is nonspecific to individual features of muscle microstructure. To address this gap in our understanding of skeletal muscle DTI’s capacity to non-invasively quantify muscle microstructure, we sought to establish an accurate and precise, DTI based tool to measure individual components of muscle microstructure. To achieve this goal we used in silico-based simulation to create a theoretical framework between DTI and key features of muscle microstructure. A precision engineered, nano-fabricated diffusion phantom was developed and tested to validate this framework in real world experiments. The main findings from these studies were the DT is most sensitive to fiber size under 60μm diameter. Using this framework to interpret the relationship between muscle microstructure and posture, Marines with smaller muscle fibers in the erector spinae muscle group have decreased lumbar lordosis, decreased lumbar extension and decreased sacral slope. Clinically and operationally, these findings are significant because they provide the framework for a non-invasive tool that can be used to measure and predict maladaptive posture in operational conditions from musculoskeletal health, which could allow clinicians to tailor rehabilitation protocols to prevent injury, and potentially be used to predict an individual’s risk for LS injury.