Evaluating the Effectiveness of Artificial Intelligence (AI)-Supported Removable Partial Denture Design Tools in...

To evaluate whether an artificial intelligence (AI)-supported learning tool can enhance students' confidence, understanding, and overall experience in removable partial denture (RPD) design. The study also explored student perceptions of AI within dental education.

A mixed-methods study was conducted involving Undergraduate Dental Technology and Prosthetics students and Undergraduate and Postgraduate Dentistry students with prior training in RPD design. Of the 40 students invited to participate, 35 completed the study. Participants independently completed conventional RPD designs before using an AI-assisted RPD design tool that generated design suggestions based on prosthodontic design principles for comparison and reflection. Pre- and post-intervention questionnaires containing Likert-type items and open-ended questions were administered using Qualtrics. Quantitative data were analyzed using non-parametric tests, Wilcoxon signed-rank test was used to assess pre- and post-survey differences within participants, while the Mann-Whitney U test was used to compare post-survey responses between Dentistry and Dental Technology and Prosthetics students and between clinical-based and laboratory-based student groups. Analyses were conducted using IBM SPSS. Qualitative responses from the open-ended questions were reviewed and grouped into four thematic categories to identify recurring perspectives and experiences related to the use of the AI-assisted RPD design tool.

Students showed significant improvements in confidence and understanding following use of the AI tool. Items related to the tool's effectiveness, design experience, and visualization received positive ratings, with median scores of 4 (interquartile range [IQR] 3-4), 4 (IQR 4-4), and 4 (IQR 4-5), respectively. Willingness to use AI and perceived helpfulness remained stable, likely due to high baseline attitudes. Qualitative findings supported the quantitative results, with participants highlighting the tool's value for visualizing RPD frameworks and guiding design decisions, while also identifying usability challenges and limitations in design flexibility.

The findings indicate that the AI-supported tool contributed positively to RPD design learning by enhancing student confidence and conceptual understanding. While it does not replace manual or clinical skill development, the tool shows clear potential as a complementary educational resource that helps bridge the gap between theoretical knowledge and practical application. Further research involving larger cohorts is recommended to confirm and expand upon these findings.