A new AI tool could help speed up the process of getting eligible Alzheimer’s patients to treatment.
The system will review medical records, flag missing information, and help prioritize referrals.
Researchers will test whether the technology can shorten the time from diagnosis to a patient’s first infusion.
For people diagnosed with Alzheimer’s disease who may qualify for newer treatments, getting from diagnosis to therapy can involve several steps — and plenty of waiting. Patients may need a referral, a specialist evaluation, brain imaging, and insurance authorization before receiving their first infusion.
Researchers at Vanderbilt Health are now working on an artificial intelligence tool designed to make part of that process move faster. The project, funded by a grant from Eli Lilly and Company, will create an AI “triage agent” that works inside Vanderbilt Health’s electronic health record system. The 18-month project is focused on the handoff from primary care or geriatrics to neurology.
“By the time a patient reaches our clinic, the clock has often been running for weeks,” researcher Amalia Peterson, M.D. said in a news release.
“This tool will be designed to make sure that when a referral arrives, we already have the information we need to act quickly. It doesn’t replace clinical judgment, but it will remove a lot of the friction that keeps patients waiting.”
How the AI will work
When a clinician submits a referral because of concerns about a patient’s cognitive health, the AI tool will review the relevant information in that patient’s chart. It will summarize the information, identify details that are missing, and recommend whether the referral should be handled as a priority or through the standard process.
Importantly, doctors and other clinicians will remain in control. They can accept, change, or override the AI’s recommendations.
Before deploying the tool, researchers will study where delays are happening now. They plan to use AI to reconstruct care timelines for more than 5,300 Vanderbilt Health patients, then work with clinicians and operational staff to identify the reasons for those delays.
The team will use what it learns to develop and test the AI agent in a pilot.
What this could mean for patients
The researchers will judge the project by one key measure: whether it reduces the amount of time between an Alzheimer’s diagnosis and a patient’s first infusion.
For eligible patients and their families, shaving time off the referral and evaluation process could make the path to treatment less cumbersome. The project is specifically focused on making sure referrals arrive with the information needed for the next step, rather than replacing doctors’ judgment.
“For many patients and families, the pathway begins in primary care or geriatrics, where cognitive concerns are first recognized, and the next steps can be difficult to navigate,” researcher Sean Huang, M.D. said in the release.
“Our goal is to help clinicians identify what information is needed earlier, streamline referral communication, and make the handoff to specialty care timelier and complete.”