Table 1 Best practices and key considerations in implementation of artificial intelligence

Validity

Choose appropriate metrics to assess model performance^12,13,14.

Evaluate whether the model achieves appropriate performance with consideration of the clinical context¹⁴.

Conduct validation studies to assess the model’s applicability to real-world clinical practice^17,18.

Usefulness

Assess the AI solution’s net benefit by weighing benefits and risks and considering workflows that mitigate risks^18,21,22,23.

Assess usefulness based on factors such as resource utilization, time savings, ease of use, workflow integration, end-user perception, alert characteristics (e.g., mode, timing, and targets), and unintended consequences^9,22,24.

Transparency and equity

Disclose information about the data and methods used to create the AI system^25,26.

Disclose which patient characteristic variables that have historically been used to discriminate are included in the model and present clear justification^21,27,28,29.

Assess model performance across key patient subgroups^10,30,31.

Assess whether the AI system is equally accessible to those who may benefit²⁵.

Provide end-users with explanations and insights about the AI system’s processes and its potential biases and errors³².

Notify patients when AI is being used and, when appropriate, to obtain their consent—particularly in sensitive or high-stakes situations^33,34.