AI in medical practice

Published: October 2024

Artificial Intelligence (AI) has emerged as a powerful technology that can potentially transform medicine. This can provide physicians with opportunities to enhance patient care and reduce administrative burden. However, health professionals must understand the risks and issues in order to make informed decisions. Items to consider include the following.

• AI is difficult to define broadly, as the term applies to a wide spectrum of applications and tools. Physicians should understand the risks and benefits of the specific AI tools that they plan to use in their practice.
• To ensure that the tools are safe and reliable, consider whether there has been regulatory approval from Health Canada; whether your College has guidelines or standards around AI; and whether a professional association or medical society has endorsed specific technologies or vendors.
• Understand the privacy and security parameters of the tool. Consider if and how the AI tool collects and uses patient data, and ensure that patients have provided informed consent regarding the collection, storage, or sharing of their health information.
• All AI models will have some inherent bias. Bias can also be present in how an AI tool is deployed, implemented, operated, and maintained. To mitigate bias, it is critical to recognize that it can and does occur, even in the absence of harmful intent.

We encourage members who have questions about using AI in their practice to review our position paper and other resources.

Video

Canadian physicians using AI: The medico-legal lens

What exactly is AI?

Knowing what we’re talking about when we’re talking about AI can help physicians understand whether a given tool really is AI.

What exactly is AI? (1:39)

Knowing what we’re talking about when we’re talking about AI can help physicians understand whether a given tool really is AI.

Not all AI has the same risk (2:20)

We can measure the risk of different AI tools by thinking in terms of two factors: the patient impact and the tool’s degree of autonomy.

Bias exists in AI (1:58)

AI can get things wrong. The data used to train AI tools may be skewed or out of date, or may contain false information.

Use case: AI scribes (1:59)

AI scribes are common in clinical practice. If you’re considering using a scribe, there are questions to ask first.

Using AI appropriately will take a village (1:47)

At CMPA, we want to see AI use evolve in a way that doesn’t place the onus entirely on physicians.

Full video: Canadian physicians using AI (10:24)

Artificial intelligence (AI) has emerged as a powerful technology that can potentially transform medicine. Health professionals must understand the risks and issues in order to make informed decisions. This video synthesizes information from our position paper, The medico-legal lens on AI use by Canadian physicians.

Good practices

• Considerations for adopting and implementing artificial intelligence (AI) in healthcare
• Clinical decision-making

Position paper

• The medico-legal lens on AI use by Canadian physicians

Articles

• AI Scribes: Answers to frequently asked questions
• The emergence of AI in healthcare
• Recording clinical encounters with patients: What physicians need to know
• What to do if you’re notified of a College complaint

Webinar

• CMPA - Navigating AI in healthcare

Research

• Legal and Ethical Challenges in Integrating AI Into Clinical Practice: Qualitative Study of Physicians’ Real-World Experiences
• Evaluating literature reviews conducted by humans versus ChatGPT: Comparative study