Page 147 - Demo

9Appendices145data. A Gaussian Naïve Bayes model was trained to classify tooth removal procedures. The study had data from 110 successful tooth removal experiments to train the classification model. Out of 75 clinically designed features, 33 were selected for the model. The overall accuracy of the classification model was 86% in the training set and 54% in the test set. The model correctly classified the jaw (upper or lower) in 95% of cases and either the correct class or a neighboring class of teeth in 88% of cases. This multiclass classification model represents a significant step towards better understanding tooth removal procedures. Despite the relatively small dataset, the quality of the data was sufficient to develop a model with reasonable performance. The results of the feature engineering, selection process, and classification model have the potential to contribute to the development of evidence-based educational material and clinical guidelines in the future.In conclusion, in the chapters of this PhD thesis various aspects of robot technology and its applications in dentistry were explored, aiming to assess existing knowledge and evidence-based practices in this field. They highlight the increasing interest in robot initiatives in dentistry, the need for scientific validation and evidence-based practices, and the potential of robot technology to increase our fundamental understanding of tooth removal procedures. The studies emphasize the importance of data collection, analysis, and collaboration between different disciplines to enhance our fundamental understanding of tooth removal procedures, with a focus on improving dental education and ultimately patient care.Tom van Riet.indd 145 26-10-2023 11:59