Browsing by Author "Powell, James N."

Now showing 1 - 2 of 2
  • Thumbnail Image
    ItemOpen Access
    The innervation of the human acetabular labrum and hip joint: an anatomic study
    (BioMed Central, 2014-02-14) Abdullah Alzaharan; Bali, Kamal; Gudena, Ravi; Railton, Pamela; Ponjevic, Dragana; Matyas, John R.; Powell, James N.
  • Thumbnail Image
    ItemOpen Access
    A quantitative computed tomography approach towards opportunistic osteoporosis screening
    (2020-03-20) Michalski, Andrew Steven; Boyd, Steven Kyle; Edwards, William Brent; Powell, James N.; Johnston, James B.; Salo, Paul T.; Manske, Sarah Lynn
    Opportunistic computed tomography (oCT) complements dual X-ray absorptiometry (DXA) by screening for osteoporosis and determining subject-specific fracture risk. Quantitative CT-based bone mineral density (BMD) and finite element (FE) estimated bone strength outcomes are known to improve fracture prediction, as compared to DXA areal BMD. However, there are shortcomings of oCT, which limit its ability to be clinically integrated as a skeletal health assessment tool for the purpose of identifying individuals at high risk of fracture that have not yet had any additional osteoporosis screening, such as a DXA scan. In this dissertation, the oCT limitation of understanding how CT scan acquisition parameters influence the skeletal health assessment is first investigated by identifying differences between CT reconstruction kernels. By using a bone-type kernel, the estimated FE failure load was increased by 18.2%, as compared to a standard-type kernel, suggesting that a standardized reconstruction kernel should be used when performing any oCT analyses. An internal density calibration method was then developed and validated to overcome the limitation of requiring a density calibration phantom within the scan field-of-view to perform oCT skeletal assessment. The developed internal calibration approach uses five reference regions and relates the known Hounsfield Units to equivalent mass attenuation values and then to equivalent bone density values. This approach was validated both in cadavers and an in vivo cohort, and it was shown to have a precision of 7.2% for skeletal health assessment outcomes. Finally, an oCT screening cohort was established using clinically acquired abdominal CT scans and was used to predict low energy fracture at known major osteoporotic fracture sites. Using this cohort, oCT screening resulted in a maximum predictive value of 0.710 for the area under the receiver operator characteristic curve to predict women with low energy fractures. These findings overcome some of the shortcomings currently preventing oCT screening from being clinically integrated. By using the millions of CT scans performed each year, oCT screening can repurpose these scans to assess skeletal health and reduce the costs and burden of fracture to both the healthcare system and society.