Marker-free registration for the accurate integration of CT images and the subject's anatomy during navigation surgery of the maxillary sinus

Abstract

Objective: This study compared three marker-free registration methods that are applicable to a navigation system that can be used for maxillary sinus surgery, and evaluated the associated errors, with the aim of determining which registration method is the most applicable for operations that require accurate navigation.

Methods: The CT digital imaging and communications in medicine (DICOM) data of ten maxillary models in DICOM files were converted into stereolithography file format. All of the ten maxillofacial models were scanned three dimensionally using a light-based three-dimensional scanner. The methods applied for registration of the maxillofacial models utilized the tooth cusp, bony landmarks and maxillary sinus anterior wall area. The errors during registration were compared between the groups.

Results: There were differences between the three registration methods in the zygoma, sinus posterior wall, molar alveolar, premolar alveolar, lateral nasal aperture and the infraorbital areas. The error was smallest using the overlay method for the anterior wall of the maxillary sinus, and the difference was statistically significant.

Conclusion: The navigation error can be minimized by conducting registration using the anterior wall of the maxillary sinus during image-guided surgery of the maxillary sinus.

Figures

Figure 1

Stereolithography (STL) files of the three-dimensionally scanned maxillofacial models and the STL files generated by converting the CT data were overlaid using the three different registration methods in the software with maxillofacial digital model images. Black arrowheads: a four-point registration method was used for the tooth cusp: the first molar mesiobuccal cusp and the mesial point of the incisal edge of the upper incisor were used as the reference points, in addition to the canine cusps on the same side as the maxillary sinus registration. Black arrows: a three-point registration method was used for the bone structures: the lateral nasal aperture, infraorbital foramen and inferior zygomaticomaxillary suture on the same side as the maxillary sinus registration. White arrow: for the surface registration method, the anterior wall of the maxillary sinus was designated as the reference region on the software, and the defined surface was used for the registration. Light grey image: STL image generated by converting the CT data. Part of image indicated by white arrowhead: STL image of the three-dimensionally scanned maxillofacial model