Depiction of pelvic fractures using 3D volumetric holography: comparison of plain X-ray and CT.

Abstract

The goal of this study was to demonstrate the feasibility and test the diagnostic performance of a new technology, 3D volumetric holography, for imaging pelvic fractures. The management of pelvic fractures may be complex, and advanced imaging studies such as CT are frequently indicated. Multiplanar CT reformations and 3D renderings provide clinically useful and complementary display of the directly acquired CT data. With the recent availability of volumetric multiple exposure holograms, produced from serial image data, it is now possible to produce true 3D images of the pelvis. In the hologram, one may view the CT data in 3D or as individual planar slices constituting the whole 3D pelvis. The diagnostic performance of the volumetric multiple exposure holograms was tested against routine radiography, CT, and 3D volumetric CT reconstructions. Routine radiography and CT were performed in 15 patients with suspected pelvic fractures. Volumetric multiple exposure holograms and 3D volumetric CT reconstructions were created from the CT data. Axial and multiplanar reformation CT images were used as the standard for fracture, diastasis, and intraarticular fragment detection. Radiograms detected 39 of 50 of the fractures and diastases and no intraarticular fragments. The 3D CT reconstructions and the holograms viewed as 3D objects alone missed two small fractures of the anterior column and one hip with intraarticular bone fragments. When the volumetric multiple exposure holograms were viewed as a 3D object and as individual planar slices constituting the whole, their results were the same as the standard. Volumetric multiple exposure holograms were as sensitive and specific as axial CT and multiplanar reformations in detecting fracture pathology. By containing and making available, from one image, both planar and 3D information, volumetric multiple exposure holograms detected subtle anatomical features that were hidden by overlapping structures in the radiographs and the 3D CT images.