The University of Washington Institute of Medical Data Science is excited to announce the winners of the 2026-27 Pilot Awards. Now in its fourth year, this initiative aims to transform medicine and enhance public health through innovative data-driven approaches.  Pilot Awards furthermore serve to seed funding to stimulate research that enhances the UW real-world data from the Electronic Medical Record and other sources of patient data.  Awardees were chosen from a wide range of applicants in fields such as radiology, computing, engineering, and medicine. 

Pilot funding for the period of July 1, 2026 to June 30, 2027, enables investigators to obtain preliminary data in order to establish a proof of concept and go on to seek larger grants. Investigators supported by the Pilot Program are working toward improving personalized chemotherapy medication preparation, towards predicting rearrest incidents following out-of-hospital cardiac arrest, as well as approaches to safer and broader access to anti-amyloid Alzheimer’s therapy.

Continue reading to learn more about their projects and join us at next year’s IMDS Symposium, where awardees will present updates on their research over the coming months.

Please join us in congratulating our awardees!

” Individualized Prediction of Amyloid-Related Imaging Abnormalities (ARIA) for Safer and Broader Access to Anti-Amyloid Alzheimer’s Therapy“​

New anti-amyloid antibody drugs are the first treatments shown to slow Alzheimer’s disease, but 20–35% of treated patients develop amyloid-related imaging abnormalities (ARIA) — brain swelling or microbleeds — and current tools predict this risk only modestly. This project will develop a machine-learning model that combines a patient’s genetics, amyloid burden, and structural brain-imaging findings with a novel measure of blood-vessel function extracted from routine MRI. By flagging high-risk patients more accurately while clearing those who are genuinely safe to treat, the work aims to make these breakthrough therapies both safer and accessible to more of the people who could benefit.

“Multimodal Biosignal Analysis for Prediction of Rearrest Following Out-of-Hospital Cardiac Arrest”​

” Computer Vision to Automate Assessment of Personalized Chemotherapy Medication Preparation to Prevent Medication Errors“​

Chemotherapy medications are the second most common class of medications associated with fatal medication errors.  They typically require multiple steps for preparation under a fume hood and are tailored to each individual patient.  This project uses artificial intelligence to assess images taken at each step of medication preparation and automatically identify medication names and volumes, which are then compared to the physician’s order.  We aim to build an automated system which can alert pharmacists to discrepancies during medication preparation, minimizing pharmacists manual image review tasks and ultimately preventing medication errors from harming the patient.