Approximately 6 million fractures occur each year in the United States, with an estimated medical and loss of productivity cost of $99 billion. As our population ages, it can only be expected that these numbers will continue to rise. While there have been recent advances in available treatments for fractures, assessment of the healing process remains a subjective process. This study aims to demonstrate the use of micro-computed tomography (microCT)-based structural rigidity analysis to accurately and quantitatively assess the progression of fracture healing over time in a rat model. The femora of rats with simulated lytic defects were injected with human BMP-2 cDNA at various time points postinjury (t = 0, 1, 5, 10 days) to accelerate fracture healing, harvested 56 days from time of injury, and subjected to microCT imaging to obtain cross-sectional data that were used to compute torsional rigidity. The specimens then underwent torsional testing to failure using a previously described pure torsional testing system. Strong correlations were found between measured torsional rigidity and computed torsional rigidity as calculated from both average (R2 = 0.63) and minimum (R2 = 0.81) structural rigidity data. While both methods were well correlated across the entire data range, minimum torsional rigidity was a better descriptor of bone strength, as seen by a higher Pearson coefficient and smaller y-intercept. These findings suggest considerable promise in the use of structural rigidity analysis of microCT data to accurately and quantitatively measure fracture-healing progression.