A Bone Fracture Risk Metric Based on Statistical Shape and Density Modeling (SSDM), 18-R8162
Principal Investigators
Todd L. Bredbenner
Daniel P. Nicolella
Robert L. Mason
Inclusive Dates: 06/29/10 – 04/04/11
Background — The problem of increased risk of skeletal fractures caused by bone mass loss in aging or disease is a major clinical problem leading to estimated health care costs of nearly $17 billion in the United States. Notwithstanding the economic burden, non-vertebral fractures, particularly of the hip, are a significant cause of morbidity and mortality in the aging population. More than 4 percent of hip fracture patients die during the initial hospitalization following fracture, and 24 percent will die within the first year. At present, clinical fracture risk assessment is most often based on areal BMD (aBMD) and accurate identification of at-risk individuals may be compromised because three-dimensional structural factors are not assessed by aBMD. The structural integrity of an individual bone in any mechanical loading environment is dependent on the bone mineral density (BMD) distribution and the macroscopic morphology of the bone, as well as complex and interrelated cortical and trabecular bone characteristics.
Approach — The goal of this research project was to generate the fundamental knowledge required to develop and implement a risk assessment tool that is easily accessible in the clinic. Innovative risk assessment methodology was developed based on statistical shape modeling methods, which have been used to describe variability in the morphology of a population of anatomical structures. Statistical shape models capture the variability of biological structures by projecting a high-dimensional representation of the structure onto a lower dimensional subspace of possible shapes constructed from a population of training shapes. In this study, researchers retrospectively investigated the use of fracture risk classifiers derived using statistical shape and density modeling (SSDM) methods, which describe both the multivariate geometry and BMD distribution variation contained implicitly within 3D imaging data for a set of bones. The performance of SSDM-based fracture classifiers in predicting fracture risk was compared to that of BMD alone.
Accomplishments — A new fracture classifier based on statistical shape and density modeling (SSDM) was developed to predict the future risk of bone fracture with significantly greater sensitivity and specificity compared to current clinical fracture risk assessment methods. This new bone fracture risk classification method will provide the basis for a revolutionary and more reliable clinical and research tool to predict future fracture risk in individuals. In addition, researchers demonstrated that the new method significantly outperforms a competing emerging method, which is based on using finite element models to compute a measure of bone strength in predicting future fracture. Using prospective data from a large National Institutes of Health clinical osteoporosis fracture study, SSDM-based fracture risk classifiers correctly classified 62.5 percent of individuals who would go on to suffer a hip fracture (fracture cases) whereas current density-based clinical methods only identified 7.5 percent of future fracture cases. Additionally, using previously collected cadaver femur data, SwRI researchers demonstrated that statistical shape and density modeling predictions of bone fracture strength are statistically equivalent to current clinical (bone density based) and emerging (finite element modeling based) methods of fracture risk assessment.
