A Novel Method for Digital Reconstruction of the Mucogingival Borderline in Optical Scans of Dental Plaster Casts

Abstract

Adequate soft-tissue dimensions have been shown to be crucial for the long-term success of dental implants. To date, there is evidence that placement of dental implants should only be conducted in an area covered with attached gingiva. Modern implant planning software does not visualize soft-tissue dimensions. This study aims to calculate the course of the mucogingival borderline (MG-BL) using statistical shape models (SSM). Visualization of the MG-BL allows the practitioner to consider the soft tissue supply during implant planning. To deploy an SSM of the MG-BL, healthy individuals were examined and the intra-oral anatomy was captured using an intra-oral scanner (IOS). The empirical anatomical data was superimposed and analyzed by principal component analysis. Using a Leave-One-Out Cross Validation (LOOCV), the prediction of the SSM was compared with the original anatomy extracted from IOS. The median error for MG-BL reconstruction was 1.06 mm (0.49-2.15 mm) and 0.81 mm (0.38-1.54 mm) for the maxilla and mandible, respectively. While this method forgoes any technical work or additional patient examination, it represents an effective and digital method for the depiction of soft-tissue dimensions. To achieve clinical applicability, a higher number of datasets has to be implemented in the SSM.

Keywords: implant planning; mucogingival borderline; soft tissue dimensions; statistical shape model; virtual planning.

Figure 1

The MG-BL can be identified by the difference in color of attached gingiva and free mucosa. The MT-BL is clearly visible at the junction between tooth and gingiva (a). The MT-BL and MG-BL are placed in 3D Slicer using curves based on sliding semi-landmarks (b).

Figure 2

Detailed depiction of the sliding semi-landmarks set along the MT-BL (blue) and MG-BL (red). To achieve a proper representation of the MT-BL the tip of each interdental papilla was annotated with one additional landmark from each side (oral and buccal).

Figure 3

Boxplot showing the error distribution of the reconstruction of the MG-BL on all tested individuals using the developed SSM.

Figure 4

Best (left) and worst (right) prediction of the MG-BL of the maxilla based on the average error.

Figure 5

Best (left) and worst (right) prediction of the MG-BL of the mandible based on the average error.