Module 2
Module 2: Structure & Specificity
Apply AI basics to clinical documentation: SOAP notes, formatting, and templates
Welcome to Module 2!
Now that you speak “AI”, let’s speak “Medicine”.
In this module, we will apply the concepts of Input -> Prompt -> Output to create clinical documentation. We will focus on Structure (SOAP notes) and Specificity (getting exactly what you want).
Learning Objectives
- Generate a basic SOAP note from a transcript
- Control formatting (lists, bullet points, order)
- Use Templates to ensure consistency
Lesson 1: The Standard SOAP Note
The most common clinical document is the SOAP note. AI is very good at this if you tell it exactly how to organize the information.
Exercise 2.1: Your First SOAP Note
You have a transcript of a hypertension follow-up. Your goal is to turn this conversation into a structured note.
Exercise 2.1: Your First SOAP Note
Your Challenge
Write a prompt that generates a standard SOAP note from the transcript.
View Patient Transcript
Generated Output:
Lesson 2: Formatting & Lists
“Make it a SOAP note” is good. “Make it a SOAP note with a numbered plan and bulleted HPI” is better.
Specificity is the key to saving time. If you have to reformat the note manually, the AI hasn’t done its job.
Exercise 2.2: Formatting Control
Let’s try a new patient (Diabetes). This time, be very specific about the format.
Exercise 2.2: Formatting & Lists
Your Challenge
Create a SOAP note where the HPI uses bullet points and the Plan is a numbered list.
View Patient Transcript
Generated Output:
Lesson 3: Using Templates
Remember “Few-Shot” prompting from Module 1? Templates are the ultimate form of Few-Shot prompting.
Instead of describing the format, you simply show the AI the empty shell you want it to fill.
Exercise 2.3: The Template Approach
Exercise 2.3: Using Templates
Your Challenge
Provide a specific template for the AI to fill in. Include placeholders like [Patient Name] and [Diagnosis].
View Patient Transcript
Generated Output:
Module 2 Complete! 🎉
You now have the tools to build solid, structured clinical notes.
Next, we’ll explore a completely different way to use AI: Refinement. Instead of generating notes from scratch, you’ll learn to fix and polish existing text.