Using Radon transform of standard radiographs of the hip to differentiate between post-menopausal women with and without fracture of the proximal femur

Assessment of risk of femoral neck fracture with radiographic texture parameters: a retrospective study

PURPOSE

To investigate whether femoral neck fracture can be predicted retrospectively on the basis of clinical radiographs by using the combined analysis of bone geometry, textural analysis of trabecular bone, and bone mineral density (BMD).

MATERIALS AND METHODS

Formal ethics committee approval was obtained for the study, and all participants gave informed written consent. Pelvic radiographs and proximal femur BMD measurements were obtained in 53 women aged 79-82 years in 2006. By 2012, 10 of these patients had experienced a low-impact femoral neck fracture. A Laplacian-based semiautomatic custom algorithm was applied to the radiographs to calculate the texture parameters along the trabecular fibers in the lower neck area for all subjects. Intra- and interobserver reproducibility was calculated by using the root mean square average coefficient of variation to evaluate the robustness of the method.

RESULTS

The best predictors of hip fracture were entropy (P = .007; reproducibility coefficient of variation < 1%), the neck-shaft angle (NSA) (P = .017), and the BMD (P = .13). For prediction of fracture, the area under the receiver operating characteristic curve was 0.753 for entropy, 0.608 for femoral neck BMD, and 0.698 for NSA. The area increased to 0.816 when entropy and NSA were combined and to 0.902 when entropy, NSA, and BMD were combined.

CONCLUSION

Textural analysis of pelvic radiographs enables discrimination of patients at risk for femoral neck fracture, and our results show the potential of this conventional imaging method to yield better prediction than that achieved with dual-energy x-ray absorptiometry-based BMD. The combination of the entropy parameter with NSA and BMD can further enhance predictive accuracy.

Trabecular homogeneity index derived from plain radiograph to evaluate bone quality

Radiographic texture analysis has been developed lately to improve the assessment of bone architecture as a determinant of bone quality. We validate here an algorithm for the evaluation of trabecular homogeneity index (HI) in the proximal femur from hip radiographs, with a focus on the impact of the principal compressive system of the trabecular bone, and evaluate its correlation with femoral strength, bone mineral density (BMD), and volumetric trabecular structure parameters. A semiautomatic custom-made algorithm was applied to calculate the HI in the femoral neck and trochanteric areas from radiographs of 178 femoral bone specimens (mean age 79.3 ± 10.4 years). Corresponding neck region was selected in CT scans to calculate volumetric parameters of trabecular structure. The site-specific BMDs were assessed from dual-energy X-ray absorptiometry (DXA), and the femoral strength was experimentally tested in side-impact configuration. Regression analysis was performed between the HI and biomechanical femoral strength, BMD, and volumetric parameters. The correlation between HI and failure load was R(2)  = 0.50; this result was improved to R(2)  = 0.58 for cervical fractures alone. The discrimination of bones with high risk of fractures (load <3000 N) was similar for HI and BMD (AUC = 0.87). Regression analysis between the HIs versus site-specific BMDs yielded R(2)  = 0.66 in neck area, R(2)  = 0.60 in trochanteric area, and an overall of R(2)  = 0.66 for the total hip. Neck HI and BMD correlated significantly with volumetric structure parameters. We present here a method to assess HI that can explain 50% of an experimental failure load and determines bones with high fracture risk with similar accuracy as BMD. The HI also had good correlation with DXA and computed tomography-derived data.

Combination of radiograph-based trabecular and geometrical parameters can discriminate cervical hip fractures from controls in individuals with BMD in non-osteoporotic range

Majority of hip fractures occur in individuals with bone mineral density (BMD) in non-osteoporotic range. This suggests that factors other than BMD are associated with increased fracture risk in these individuals. The aim of this study was to investigate the combined ability of radiograph-based trabecular and geometrical parameters to discriminate cervical hip fractures from controls in individuals with non-osteoporotic BMD. A total of 39 postmenopausal females with non-pathologic cervical hip fracture were recruited to the study. Nineteen of the fracture patients (48.7%) had non-osteoporotic BMD and they constituted the fracture group. The control group consisted of 35 BMD-matched non-osteoporotic females. Several geometrical and trabecular parameters were extracted from plain pelvic radiographs, and their combined ability to discriminate fracture patients from controls was studied using a receiver operating characteristics (ROC) analysis. Significant differences in several radiograph-based geometrical and trabecular parameters were found between the fracture patients and controls, whereas no statistically significant difference in BMD was observed (p=0.92) between the groups. Area under the ROC curve was 0.993 (95% CI 0.977-1.008) for the combined multiple regression model, which included both trabecular and geometrical parameters as explanatory factors. Here, the sensitivity of 100% was achieved with the specificity of 94%. In a cross-validation of the model, 94.4% of the fracture patients, and 94.1% of the controls were classified correctly. The combination of radiograph-based trabecular and geometrical parameters was able to discriminate the cervical hip fracture cases from controls with similar BMD, showing that the method can provide additional information on bone structure and fracture risk beyond BMD.