Why PIs and Researchers Need Brim

You’re not reading EHRs for fun — you’re working to uncover real insights that could improve medical outcomes. Maybe you’re studying how to repurpose an existing drug, mapping correlations between pickleball and orthopedic injuries, or identifying best practices in emergency general surgery.

You’re a principal investigator, a medical student, a researcher with an MD or PhD. And what you really want is to get to the analysis — not to spend your day lost in a maze of medical records.

But here’s the reality:
- The records are long. Each one a hundred notes, and often a lot more.
- They’re also messy — clinical notes mixed with discharge summaries, radiology reports, problem lists, and clinician communications. They’re repetitive and inconsistent.
- And for your study, you only need a few crucial data points per patient.

So what happens?

You spend hours going case by case. One screen has the EHR open, the other has REDCap and the PDF of your protocol. Maybe there’s a spreadsheet. Maybe a pen and paper to keep track of the timeline. You’re manually abstracting patient data — one field at a time.

Sound familiar? You need Brim.

‍

Brim Makes Chart Abstraction Easier — and Smarter

‍

Brim is an AI chart abstraction tool designed for medical researchers like you. It's built to simplify the abstraction phase of conducting retrospective research, so you can focus on analysis and discovery.

Start by importing your cohort of medical records directly into Brim (it's fully secure - more on that below). Then, define what you want to abstract. You can pull in your protocol straight from REDCap, upload a CSV, or define fields manually.

Brim helps you build and tweak your variables based on what you want to abstract.

Here's where the magic happens:

1. Choose how many patients you want to process. Hit “Generate.” 
2. Go grab a coffee (or a tea, we won't judge).
3. By the time you’re back, Brim has drafted the abstraction — complete with citations from the raw note(s) informing each data point. 

Then you become the editor: your job is to review and refine, giving feedback so Brim can learn and improve the next batch of patients.

Even better: if your analysis raises new questions — say tennis injuries start showing up in your pickleball study — it’s easy to update or add some fields and re-abstract. The abstracted fields can flex and expand as you learn more about your data.

When you're done, export everything to CSV. The data is clean, curated, and ready to support your publication or grant.

‍

Seriously, Why Are You Still Doing This Manually?

‍

Medical research deserves better than copy-paste. With Brim, your work moves faster, your insights are sharper, and your time is spent where it matters.

And right now, we’re offering a limited-time free license to the first team at any institution to adopt Brim. Get in touch here.

‍

For Your Security Team:

Brim is fully containerized and can run inside your organization’s Azure or AWS environment. It also connects to your institutionally-approved large language model (LLM) endpoint, so your patient data stays within your control. We'd love to talk to your security team about it.

‍