Third-year Feinberg student Tyler Smith has worn many hats. He’s a self-proclaimed “math nerd,” former collegiate cross country and track athlete and former AmeriCorps member. Smith never envisioned himself as an entrepreneur though.

One month ago Smith attended a national medical innovation competition in Washington, D.C. where he pitched Lipid Llama, a medical artificial intelligence chatbot, to a panel of three judges.

“Being in front of venture capitalists and physician entrepreneurs was a very different world,” Smith said.

Smith co-created Lipid Llama in January 2024 as a project for his Masters of Science in Artificial Intelligence capstone class with Jonathan Hourmozdi, a Feinberg cardiovascular fellow, and Amogh Karnik, now a Feinberg cardiology faculty member. 

Smith said the students wanted to create a product that strengthened health literacy and helped patients understand their medical tests.

Smith said the group focused its creation on lipid panels, a common blood test that monitors risk of cardiovascular disease, along with cholesterol and triglyceride levels. Users can input their test results, and the app will interpret these results and answer questions about each piece of data.

The app’s name refers to the blood test and a family of large language models called Llama, built by Meta AI.

Although lipid panels are one of the most popular medical tests, many patients have difficulty understanding their results and don’t ask their doctors for clarification, Smith said.

“Lipid tests carry some really jargon-heavy terms, like HDL cholesterol and LDL cholesterol,” Smith said. “These are things that patients hear, and they go in one ear and out the other.”

Lipid Llama uses a risk score calculator from the American Heart Association that analyzes a patient’s cholesterol results, blood pressure and medications in order to provide a 10-year risk assessment of heart attack, stroke and heart disease.

Smith said he hopes to pilot Lipid Llama with patients locally once he secures additional funding for the app and approval from the Institutional Review Board, an administrative body that monitors biomedical research with human subjects.

In April, Smith attended Innovation Festival, the American Medical Student Association’s  inaugural digital health technology competition. Lipid Llama was one of eight finalists selected from over 30 teams.

While Smith didn’t take home first place, he said the competition gave him the opportunity to “pick the brains” of several physician entrepreneurs.

“Coming away from (InnoFest), I learned there are a lot of avenues to be successful,” Smith said. “I learned that leaning on the expertise of your team is really important.”

Hourmozdi, Karnik and Smith, the only three Feinberg students in the MSAI program, all shared a passion for cardiovascular medicine. 

In addition to providing medical analysis, Hourmozdi said Lipid Llama’s chatbot feature encourages patients to play an active part in their heart health. Users can ask questions about their test results to better inform themselves about steps toward preventing heart-related problems, he said. 

“There’s a lot of nuance and a lot of misinformation that’s out there about the interpretation of lipid panels,” Karnik said. “We wanted to make sure that we could address that need, while also developing something that we would use in our own practice.”

After their capstone class ended in the spring of 2024, Hourmozdi said Smith took the lead on the next phase of Lipid Llama’s development. The team designed a user-friendly interface and made Lipid Llama multilingual to expand its outreach to non-English speaking communities, Smith said. 

Smith said he is also working to publish an academic paper on Lipid Llama, which is currently pending acceptance. He wrote the paper with Hourmozdi, Kanrik, James D. Thomas, a Feinberg cardiology professor, and Adrienne Kline, the paper’s principal investigator and head of the cardiovascular AI research lab where Smith is currently a fellow.

Hourmozdi and Karnik still advise Smith in the health-related parts of Lipid Llama. Hourmozdi said Smith shadowed Karnik in a clinic to gain knowledge about cardiovascular diseases and the clinical application of biotechnology. 

Hourmozdi said Smith’s ambition with Lipid Llama has also motivated him and Karnik. 

“It has been an inspiration to see how Tyler is able to take something from ideation and really put it out on a national stage and make a real, utilizable tool that Amogh and I can bring to patients in a few short steps,” Hourmozdi said. 

MSAI Program Director Kris Hammond taught Smith’s capstone class and has mentored him throughout his app development.

Hammond said he repeatedly tells MSAI capstone students like Smith to tailor their products to their stakeholders. Lipid Llama is a prime example of a user-focused technology designed with the goal for implementation in the real world, he said. 

“There’s a culture in computation that sometimes you spend a lot of time with your head down, looking at the screen and typing away,” Hammond said. “You also need to lift your head up and look out beyond the screen and understand that what you are doing is going to have an impact on the world.”

He said the MSAI program teaches students like Smith to be “ambassadors and leaders” of technology within their respective industries. In the medical field, clinicians are already stretched thin, Hammond said, so AI tools can have a significant impact in improving health outcomes.

Smith hopes clinicians and patients in the medical field will embrace AI technologies like Lipid Llama to improve patients’ education and health outcomes.

“There’s a lot of opportunity for equity and access, and that’s what we’re focused on here,” Smith said. “It’s getting the most amount of patients the tools that they need to have access to their preventive cardiovascular health and be able to understand and ask questions. I think it’s going to do a lot of good, and I hope to help spread it.”

Email: [email protected]

X: @rschlueter26

Related Stories:

— ‘One-stop-shop’ for med students: Feinberg students create AI study tool

— Institute for Artificial Intelligence in Medicine focuses on ethical data use

— Northwestern Football alumni create AI research start-up Consensus