[Virtual rendering techniques in otologic imaging]

Virtual postprocessing techniques combine the advantages of condensing the large amounts of data provided by high-resolution (HR) cross-sectional imaging modalities with those of three-dimensional (3D) imaging. The techniques and indications for virtual representations in imaging of the middle ear (ME), internal ear (IE), and cerebellopontine angle (CPA) are presented together with practical examples.

HR data sets acquired by computed tomography (CT) and magnetic resonance imaging (MRI) in patients with ME, IE, and CPA pathologies were transferred to a workstation via an internal network to generate endo- or extraluminal 3D views by means of the volume rendering technique (VRT). The source data were acquired using scanners and imaging protocols with the highest resolution available at present: a multislice spiral CT (MSCT) with a slice thickness of 0.5 mm and a reconstruction increment of 0.2 mm and a 3D CISS sequence with a slice thickness of 0.5 mm for MRI.

Virtual endoscopy was superior to cross-sectional images for assessing ME pathologies like dysplasia, postoperative changes, and destructive bone processes with extensive soft-tissue involvement; fibrous obliterations of the internal ear and labyrinthine dysplasia were depicted with a superior image quality on 3D renderings compared to conventional reconstruction techniques. Virtual endoscopy of the CPA and external acoustic meatus (EAM) was helpful in detecting and visualizing neurovascular conflicts and in assigning small intrameatal tumors to components of the acousticofacial bundle. A common feature of all applications was that the large numbers of source images could be reduced to a few 3D reconstructions for documentation and optimized communication of the findings between the radiologist and otologist.

Virtual rendering makes an important contribution towards establishing, presenting, and documenting the findings when certain otologic pathologies have to be assessed. It can be used for routine imaging and allows for more efficient handling of the large amounts of imaging data generated by high-resolution cross-sectional imaging modalities.