At the University of Florida College of Medicine‘s Annual Medical Student Research Program (MSRP) Symposium on March 6, Intelligent Critical Care Center (IC3) medical student Anjay Shah received a Semifinalist Award and $100 for his work.
Out of 120 total projects entered by medical students from the class of 2025, Shah’s project, “Phenotyping Acute Kidney Injury Using Consensus Clustering,” was honored as one of 21 semifinalists that just missed being named one of the six finalists.
As an MSRP student, Shah received mentoring from IC3 faculty including director Azra Bihorac, associate director of research Tezcan Ozrazgat-Baslanti, assistant scientist Yuanfang Ren, and postdoctoral associate Esra Adiyeke.
“I’m really grateful for this opportunity to work with such a wonderful team and group of mentors,” Shah said. “This was my first exposure to machine learning and its clinical utility in medicine, so I’m excited to see what this team develops in the future.”
The project used consensus clustering to classify acute kidney injury (AKI) patients into distinct categories, or phenotypes, based on their readily available electronic health records (EHR). These phenotypes would help clinicians to identify the disease trajectories of AKI patients and then prescribe the optimal treatments.
After examining EHR data for 25,887 UF Health patients from 2014 to 2019, the AI consensus clustering AI model defined two unique phenotypes in a heterogenous cohort of AKI patients, and both phenotypes had significant differences in lab values, demographic data, and survival outcomes.
The next steps in this research would be to increase the generalizability of the findings by expanding the cohort size, conduct prospective studies to evaluate the impact of treatments on patient outcomes, and incorporate longitudinal data and follow-up to assess the status of the phenotypes over time.
In total at the MSRP Symposium, six finalists and 21 semifinalists received $4,700 in awards.
For more information on the UF College of Medicine’s MSRP program, click here.