SUMMARY
Yemi Omotoso, M.B.B.S., M.D., is an emergency medicine clinician-investigator who develops and evaluates artificial intelligence (AI)-based decision support tools for emergency department triage and early risk assessment. His research focuses on reducing undertriage among patients with complex medical conditions, including people with cancer and transplant recipients, by combining presenting symptoms and vital signs with information from the electronic health record.
Dr. Omotoso designs context-aware triage tools that help identify patients at increased risk of serious illness by incorporating factors such as medical history, medications and recent healthcare encounters. His goal is to improve timely access to care and reduce the risk of clinical deterioration while patients await evaluation. He also studies the implementation and validation of real-time clinical decision support systems, with an emphasis on usability, workflow integration and patient outcomes.
Focus areas
- Context-aware triage modeling. Dr. Omotoso develops machine learning models that combine triage inputs with prior diagnoses, medications and recent encounters to estimate near-term clinical risk. He focuses on improving early identification of high-acuity presentations that can appear deceptively stable on arrival, especially in patients with complex specialty care histories.
- Epic-integrated decision support. Dr. Omotoso studies how risk scores and explanatory flags can be delivered within electronic health record workflows at the triage workstation. He evaluates design choices that minimize alert fatigue, preserve clinician autonomy, and enable rapid review by nurses and charge teams.
- Prospective evaluation and safety. Dr. Omotoso designs validation studies that compare AI-assisted triage outputs with standard triage processes and downstream clinical outcomes. He emphasizes safety constraints, monitoring plans and physician-in-the-loop governance to prevent harmful misclassification and to document real-world performance.
- Care pathway optimization. Dr. Omotoso investigates how early risk stratification can support appropriate routing decisions, including identifying patients who may be suitable candidates for hospital-at-home programs. He links triage predictions to operational measures such as patient throughput, escalation events and short-term revisits.
Significance to patient care
Emergency departments are very busy, and some patients who are seriously ill may look stable at first and end up waiting too long. Dr. Omotoso's work aims to help nurses and other healthcare professionals identify patients at high risk earlier by using information already in the electronic health record, such as past illnesses, medications and recent hospital visits.
When healthcare professionals recognize these patients sooner, patients can receive care more quickly. Then symptoms or complications may be addressed before they worsen. His research also helps support safe decisions about which patients may be able to recover at home with close monitoring instead of staying in the hospital.
Professional highlights
- Gerstner Scholars Program in AI Translation, Mayo Clinic, 2026-2028.
- Dean's Scholarship, Johns Hopkins Carey Business School, Johns Hopkins University, 2018-2020.
- Lincoln Medical Center:
- Best Research Project, Annual Resident Research Symposium, 2019.
- Intern of the Year, Department of Emergency Medicine, 2019.