Synchrotron radiation and x-ray microtomography based on absorption contrast (performed at HASYLAB at DESY/Hamburg and BAM/Berlin) have been used for imaging of temporal bones and various internal components in situ at spatial resolution down to 7 micrometers with potential enhancement into the submicron range. Due to the volume imaging approach, several hidden structures (e.g., intra-ossicular channels) were revealed. Using several 3D-image processing techniques all data have been segmented into objects (e.g., bony ossicles, ligaments, fluids, air spaces) and subsequently transformed into vectorized data models. Because they are based on the original voxel resolution their content of vector primitives (e.g., polygons) is huge compared to recent models. Therefore they became polygon-reduced to fit into current computation limitations. So far individual data models of the entire hearing apparatus from tympanic membrane to cochlea out of intact specimen, including separate models of ossicles, ligaments and other components have been obtained, provided, in interchangeable data formats (e.g. vector-based: IGES, STL, VRML) and introduced into FEA for modeling of acousto-mechanic transfer characteristics of the middle ear. Their pseudo and real 3D- visualizations (rendering, autosteroscopic display, enlarged solid models) allow easy understanding of the anatomy and pathology of the human hearing organ and may support patient and student education in the field of otology and audiology.
|