Landmarks Identification Using Three Dimensional Volume Rendering and Multi-Planar Reconstruction: Is it Comparable?

(Pages: 42-48)

Shereef Shahen1, Marina Porcelli2, Samir Abu Obaida3, Rossella Carrino1, Maddalena Vitale1 and Vincenzo Grassia1

1Department of Orthodontics, University of Campania “Luigi Vanvitelli”, Naples, Italy; 2Department of Biochemistry, Biophysics and General Pathology, University of Campania “Luigi Vanvitelli”, Naples, Italy; 3Department of Orthodontics, Cairo university, Medical director, Dubai, UAE


Materials and Methods: Twenty CBCT records were collected. Three observers digitized manually twenty landmarks on each of them. In MPR, all the points were checked on the three slice directions at the same time while iso-surface appears as a solid 3D object. Identification was carried out three times at weekly intervals. M and SD were calculated for each coordinate and direct distance.
Results: Between the two viewing methods, marked differences were found for the following points: Orbitale, B, A, Menton, Gnathion, Pogonion, Infra and Supra dentale, Gonion, Porion, and Sella points. SD were greater than one mm. Moreover, some points as Porion expired as SD were bigger than two mm in both methods. Instead, B, Infradentale and Basion points expired only in iso-surface while Orbitale, Gonion, PNS and Frontozygomatic points in MPR. Moreover, for each point, one coordinate could be accurate such as Z coordinate for ANS, Orbitale, Porion, Basion, PNS, Menton, Infra and Supra dentale points, Y coordinate for B, A, ANS, Pogonion, Nasion, Infra and Supra dentale points and X coordinate for ANS and PNS points. High error points may still have reliable coordinates components.
Conclusions: The two methods for three dimensional landmarks manual identification resulted not comparable. Error was greater in MPR than iso-surface rendering and some coordinates of high error points could be still reliable.

Keywords: Three dimensional, Landmarks, Comparability, Multi-Planar Reconstruction, Iso-surface.