Clinical medical research uses something called a TFL as a way to represent statistics based on key characteristics of the subjects in a study.

Traditionally, SAS and MS Office (Word) have been utilized to create these, but more recently (and perhaps after FDA approval) R and Shiny are being considered.

In this session, I'd like to share some of the architecture, some of the features, some of the challenges and things we've tried along the way, and finally some of the lessons learned in creating TFL Builder.

More specifically, I'd like to share three special experiences I had while creating the application.

1. creating a Shiny application and using "shiny.fluent" for UI.
2. I utilized "reactable" to implement CRUD.
3. I used "temporary file" to add dynamic bookmark feature.

After completing the project, in retrospect, I wish I had prepared shiny beforehand.

1. Consider the user's scenario
2. Actively utilize shiny test
3. Keep researching on the technical things