Skip to main content

You received your blood test results. Now what?

Northwestern med student’s AI invention acts as an extension of the doctor’s office, and it’s up for a national prize
Lipid Llama provides an extension of the discussion you’ve already had with your doctor, according to its creator Tyler Smith. His invention is one of eight finalists in the American Medical Student Association competition.

Northwestern medical student Tyler Smith has a cool new invention, and he’s taking it to the finals of InnoFest, a new national competition hosted by the American Medical Student Association (AMSA) that showcases innovations from future physicians.

Smith’s invention, Lipid Llama, is one of eight finalists selected from more than 30 teams.

What is Lipid Llama?

A multilingual AI chatbot app, it’s designed to help patients better understand their lipid panel — a common blood test that measures different types of fats, including cholesterol and triglycerides, to determine the risk of heart attack, stroke and other cardiovascular diseases.

Patients can ask Lipid Llama questions based on their test results. The chatbot’s answers draw from the PREVENT calculator developed by the American Heart Association.

How does it help?

The app serves as an extension of the doctor’s office, especially for patients who leave appointments with lingering questions.

“There are so many tests, especially in the primary care setting, that when patients receive their lipid-panel results, they may have only limited time to discuss it with their doctor,” said Smith, a Chicago-area native and third-year student at Northwestern University Feinberg School of Medicine who holds a master’s degree in Artificial Intelligence from Northwestern.

“There are millions of lipid tests run every single year,” Smith said. “And while it’s a popular test, it’s also a low-risk one. We’re not going to run the risk of telling somebody with the help of AI that they have cancer or a significant disease.”

What about the name?

“Explaining the joke is always the worst part,” he said with a smile. “In the machine learning community, there’s a family of large language models called Llama models from Meta AI. It’s an inside joke that gives us a fun avatar.”

Dr. Adrienne Kline, a professor — and Smith’s mentor — praised the innovation and the innovator.

“It’s always so rewarding to see young minds taking up this field and bringing forward new ideas,” said Kline, a research assistant professor of cardiac surgery at Feinberg. “Tyler’s work is a perfect example of how medical students can shape the future of patient care.”

After a virtual pitch to health care and tech experts in February, Smith qualified for the finals, which take place Friday, April 18. What’s at stake is a potential $125,000 in venture capital funding.

“That’s more money than I’ve ever seen in my life,” Smith said. “But this means a tremendous amount because it’s the biggest project of my career. And we want to see this through.”