IADR Abstract Archives
Augmenting Dental School Instruction and Learning Outcomes Using Artificial Intelligence
Objectives: We investigate how AI models can be used to improve learning outcomes, course development, and students' perceptions in dental school courses.
Methods: Publicly available multi-modal (text, visual, audio) large language models (MLLMs) permit students to input course materials and receive learning-oriented outputs including personalized tutoring (based on course session objectives), summaries, detailed explanations, and self-testing. A sample size of forty students taking a second-year periodontics course implemented MLLMs in three ways. The first method was an input prompt designed to have the AI tutor students on specific learning objectives. The second method utilized MLLMs to generate creative ways to convey complex concepts in an interactive classroom session. The third method was uploading lecture material into a “CustomGPT” to create a chatbot trained on course data that provides more accurate and relevant outputs. Students’ perceptions will be measured subjectively by pre and post course surveys. Learning outcomes, defined as the retention and application of knowledge, will be measured objectively by exam and quiz scores. Functionality will be assessed by comparing outcomes to previous years’ course scores prior to the integration of AI.
Results: A pre-course survey of students enrolled in a second-year periodontics course revealed 63% of students already used AI to clarify unclear topics and 68% had a positive perception of AI. Post-course surveys and outcomes will be measured and analyzed at the conclusion of the course.
Conclusions: As the capacities of generative artificial intelligence models advance and adoption of the technology increases, there is potential to reshape education. The utilization of AI technology in dental school courses warrants further research that will enable faculty, educators, and students to enhance the learning experience and incorporate individualized instruction. We believe these new tools will elevate the delivery, retention, and application of dental school knowledge.
Methods: Publicly available multi-modal (text, visual, audio) large language models (MLLMs) permit students to input course materials and receive learning-oriented outputs including personalized tutoring (based on course session objectives), summaries, detailed explanations, and self-testing. A sample size of forty students taking a second-year periodontics course implemented MLLMs in three ways. The first method was an input prompt designed to have the AI tutor students on specific learning objectives. The second method utilized MLLMs to generate creative ways to convey complex concepts in an interactive classroom session. The third method was uploading lecture material into a “CustomGPT” to create a chatbot trained on course data that provides more accurate and relevant outputs. Students’ perceptions will be measured subjectively by pre and post course surveys. Learning outcomes, defined as the retention and application of knowledge, will be measured objectively by exam and quiz scores. Functionality will be assessed by comparing outcomes to previous years’ course scores prior to the integration of AI.
Results: A pre-course survey of students enrolled in a second-year periodontics course revealed 63% of students already used AI to clarify unclear topics and 68% had a positive perception of AI. Post-course surveys and outcomes will be measured and analyzed at the conclusion of the course.
Conclusions: As the capacities of generative artificial intelligence models advance and adoption of the technology increases, there is potential to reshape education. The utilization of AI technology in dental school courses warrants further research that will enable faculty, educators, and students to enhance the learning experience and incorporate individualized instruction. We believe these new tools will elevate the delivery, retention, and application of dental school knowledge.