Exploring the Impact of GitHub Copilot on Health Informatics Education

Area of Science:

• Health Informatics
• Computer Science Education
• Artificial Intelligence

Background:

• Artificial intelligence (AI)-driven code completion tools, like GitHub Copilot integrated with Visual Studio, present significant implications for Health Informatics education.
• These tools are particularly relevant for students learning programming languages such as SQL and Python within this field.

Purpose of the Study:

• To evaluate the effectiveness of AI code completion tools in solving Health Informatics coursework problems.
• To assess the performance of GitHub Copilot across a spectrum of problem complexities, from simple to advanced.

Main Methods:

• Assessed GitHub Copilot's code generation capabilities for introductory Health Informatics programming course problems.
• Evaluated the correctness of generated code based on provided problem statements.
• Investigated the impact of detailed explanations on the tool's problem-solving effectiveness.

Main Results:

• GitHub Copilot successfully generated correct code for basic programming problems.
• Code correctness and effectiveness diminished as problem complexity increased.
• While the tool struggled with highly complex problems, performance improved with more detailed input explanations.

Conclusions:

• AI code completion tools are relevant for Health Informatics programming education, but require critical student evaluation.
• Educators must adapt to the rapid advancements in AI programming assistance tools.