Structuring Laboratory Classes of Artificial Intelligence in Medicine

Area of Science:

• Medical Education
• Artificial Intelligence in Medicine
• Curriculum Design

Background:

• First-year medical students often lack foundational knowledge in computer science and mathematics.
• Integrating artificial intelligence (AI) into medical education requires careful curriculum design and pedagogical approaches.
• Previous AI in medicine courses highlighted challenges in logical reasoning and project variability among students.

Purpose of the Study:

• To describe the design, implementation, and restructuring of an AI in Medicine course for first-year medical students.
• To evaluate the impact of curriculum modifications on student learning and project outcomes.
• To support the feasibility of early AI integration in medical training.

Main Methods:

• Two-year iterative curriculum development involving seminars, hands-on practical work (regressions, neural networks), and video demonstrations.
• Introduction of ontology building, comparative analysis of neural networks and regression, and prompt engineering in the second year.
• Focus on applied learning activities and guided practical exercises.

Main Results:

• Initial course highlighted difficulties in logic and ontologies, with high variability in project quality.
• Second-year adjustments, emphasizing applied work and prompt engineering, led to a significant increase in performance.
• Enhanced consistency and quality of student projects were observed after curriculum restructuring.

Conclusions:

• Early introduction of AI in medicine for medical students is feasible and beneficial.
• Iterative curriculum design, focusing on practical application and conversational AI skills, is crucial for effective AI medical education.
• The study supports the value of adapting teaching methods based on student feedback and performance data.