[P-129] Quantitative Evaluation of Digital Sculpting for Dental Anatomy Education
[Abstract]
[Objective]
CAD/CAM technology in dentistry has become an essential component in dental treatment and, consequently, in dental education. To meet this demand, Niigata University offers a training program covering dental sculpting using both conventional and digital approaches. However, students’ skill assessment, even of digital products, remains challenging. In particular, the mean and standard deviation (SD) of volumetric measurements have been used to assess product accuracy, however, substantial variability of the products can bias the assessment. In this study, we employ more reliable parameters such as statistical mode, which indicates the distribution tendency of the differences, and Interquartile Range (IQR), which represents the dispersion range of the differences in the product. The aim of this study was to assess the students’ products by using statistical mode and IQR and compare them with the conventional mean and SD.
[Method]
Three-dimensional tooth models (C12-AT1A 32S, Nissin) used in the conventional wax sculpting class were digitalized to serve as reference models. The detailed structure of the models was smoothed/flattened and used as base models for sculpting. Students then downloaded base models, digitally sculpted them using SculptGL software, and submitted the resulting products. The distance difference of each model compared to the reference model was calculated with MeshLab software. The mean, SD, statistical mode, and IQR were determined, and correlation analysis was performed between each parameter.
[Results and Discussion]
The results demonstrated that both statistical mode and mean values fell below 0, indicating that the students’ products tended to be smaller than the reference model. The statistical mode tended to show smaller values compared to the mean, and there was no correlation between these parameters (R2=0.20). The IQR and SD showed only a low correlation (R2=0.43), indicating the independence of these parameters. These results suggest that the combination of the mode and IQR may provide a more reliable assessment of student performance compared to the mean and SD. However, the relationship between the overall measurements used in this study and specific evaluation remains to be determined.
[Objective]
CAD/CAM technology in dentistry has become an essential component in dental treatment and, consequently, in dental education. To meet this demand, Niigata University offers a training program covering dental sculpting using both conventional and digital approaches. However, students’ skill assessment, even of digital products, remains challenging. In particular, the mean and standard deviation (SD) of volumetric measurements have been used to assess product accuracy, however, substantial variability of the products can bias the assessment. In this study, we employ more reliable parameters such as statistical mode, which indicates the distribution tendency of the differences, and Interquartile Range (IQR), which represents the dispersion range of the differences in the product. The aim of this study was to assess the students’ products by using statistical mode and IQR and compare them with the conventional mean and SD.
[Method]
Three-dimensional tooth models (C12-AT1A 32S, Nissin) used in the conventional wax sculpting class were digitalized to serve as reference models. The detailed structure of the models was smoothed/flattened and used as base models for sculpting. Students then downloaded base models, digitally sculpted them using SculptGL software, and submitted the resulting products. The distance difference of each model compared to the reference model was calculated with MeshLab software. The mean, SD, statistical mode, and IQR were determined, and correlation analysis was performed between each parameter.
[Results and Discussion]
The results demonstrated that both statistical mode and mean values fell below 0, indicating that the students’ products tended to be smaller than the reference model. The statistical mode tended to show smaller values compared to the mean, and there was no correlation between these parameters (R2=0.20). The IQR and SD showed only a low correlation (R2=0.43), indicating the independence of these parameters. These results suggest that the combination of the mode and IQR may provide a more reliable assessment of student performance compared to the mean and SD. However, the relationship between the overall measurements used in this study and specific evaluation remains to be determined.