Search for a command to run...
Artificial intelligence (AI) is already fundamentally changing society, with medicine being no exception. AI will impact how we practice, how hospitals operate, and even the practice of medicine itself. The use of AI-based products has already begun, with examples including AI scribes and large language models such as ChatGPT. Work is ongoing to produce models that have specific functions within medicine, such as kidney injury prediction. However, transformative foundational work, such as AlphaFold (for protein structure prediction), also promises to completely change the way we approach medicine. Therefore, clinicians must develop a clear understanding of AI, not as an optional skill, but as a core competency of modern medical practice. This paper serves as a tutorial to guide medical professionals through the basic principles of AI. It will teach clinicians how to build a mental scaffold to understand and springboard into AI. The core parts of this paper are organized in steps, with additional relevant topics addressed in modules at the end of the paper. The core steps are meant to be read sequentially. To prepare the reader for the rest of the paper, this tutorial will first introduce what AI is and then cover some basic definitions needed to understand other concepts. The reader will then be ready to understand what deep learning is and the difference between supervised and unsupervised learning. Finally, the reader will go through how deep learning models learn. Separate modules on safety and clinical applications are also included. This tutorial is relevant to clinicians at all levels but may be particularly useful for practicing clinicians who are encountering AI tools integrated into their practices without previous formal education in the field. Users of this tutorial can refer to specific sections or read the entire paper.