In the last several years, pre-collapse detection of femoral head osteonecrosis (ON) by magnetic resonance imaging (MRI) has become ever more frequent. Delay or prevention of mechanical collapse of the involved femoral head - certainly preferable to arthroplasty in this relatively young patient group - is becoming an increasingly feasible treatment goal. However, the efficacy of current therapeutic procedures in forestalling head collapse remains highly variable. We hypothesize that much of the variability in disease progression can be explained in terms of quantifiable differences in the structural competence of the involved femoral head. To test this hypothesis, we propose to study the relationship between structural deficit and clinical progression in a representative series of ON patients for whom the lesion involvement pattern can be reliably inferred from MRI. We will pursue three specific aims. First, we will further explore the relationship between MRI appearance, histologic tissue status, and mechanical property deficits in a series of resected femoral head specimens. Second, we will use an existing three-dimensional finite element model (13,192 degrees of freedom) to estimate mechanical stress distributions (based on MR imaging) in the femoral heads of 25 of our institution's osteonecrosis patients. In these patients, we will compute an effective mechanical risk factor, and we will correlate that risk factor with clinically-observed disease progression (i.e., collapse). Third, we will perform a series of in vitro benchtop experiments to physically validate the finite element model. If supported by the results of the MRI/FEM patient series (Aims 2), our hypothesis has an important clinical implication: the likelihood of disease progression in a specific patient can be predicted based on the mechanical challenge presented by that patient's particular distribution of necrosis. Should this prove to be the case, future studies can be designed to explore the efficacy of alternative intervention strategies in reducing that mechanical challenge, on a patient-specific basis.