Write an annotated bibliography:Advances in artificial intelligence $($AI$),$ and its subfield machine learning $($ML$),$ can be seen in

almost every domain of life, including cutting$-$edge health research$\mathrm{.}1,2$ However, only a tiny

fraction of health AI$/$ML systems described in research papers makes its way into clinical

practice. To help address this issue the Hospital for Sick Children $($SickKids$)$ and the Vector

Institute for Artificial Intelligence $($Vector$)$ organized the Vector$-$SickKids Health AI Deployment

Symposium on October $30,2019$ attended by $166$ clinicians, computer scientists, policy makers,

and healthcare administrators. The aim was to showcase real$-$world examples of AI moving

from the research lab to the clinic. Speakers came from a variety of Canadian and US

institutions including St$.$ Michael$$s Hospital, the University Health Network, the University of

Waterloo, Public Health Ontario, Ontario Tech University, the University of Michigan, Northern

California Kaiser Permanente, Johns Hopkins University, University of Pennsylvania, and Duke

University. The successes and challenges that each project experienced provided valuable

insights into the new and evolving field of AI for health. Each speaker was asked to prepare a

structured presentation which touched upon the following topics:

$$ Prerequisites for deployment $($such as data access$)$

$$ AI Applications

$$ Evaluation procedures

$$ Visualization strategies

$$ Team building

$$ Ethical considerations

$$ Deployment pipeline

$$ Lessons learned

The focus was on identifying concrete $$do$$s and don$$ts$$ for deploying ML into healthcare.

The deployment strategies for ML in healthcare are currently ad hoc for most applications. A

lack of defined rules and best practices leads to provisional solutions that may be suboptimal.

Through a better knowledge of real$-$world implementations several common themes surfaced.

These examples are a first step in understanding what is required to define pathways for AI$/$ML

deployment. Academic researchers in the field of health AI$/$ML generally focus on factors

required for successful data science ranging from statistical analysis and ML algorithms to

database access and institutional review board approval. Implementation science is an equally

important discipline that is needed to bring ML tools to the bedside$\mathrm{.}3$ This field of research is by

no means new, and has been studied extensively in the context of translating medical research

into clinical practice$\mathrm{.}4$ In a related vein, understanding how and why institutions adopt and

maintain technologies is also an important component of any healthcare technology and change. During model development, in order to choose algorithms, features, and evaluation metrics that will lead to

robust and institutionally appropriate systems, research teams need to take the entire pipeline of project development into view