Advances in technology have changed the face of global healthcare through the growing use of AI, deep learning, machine learning, IoT, big data, and prescriptive analytics. These capabilities are reshaping how health systems monitor patients, support diagnosis, and allocate limited resources more effectively—especially in high-stakes, high-variability environments such as emergency care. Our recent work reflects this shift through multiple complementary directions: developing deep learning forecasting models for emergency department demand (including pediatric ED arrivals and COVID-era arrival patterns), and building predictive models that support operational decisions such as anticipating emergency department length of stay and identifying key indicators of performance during medical surges.

In parallel, we design data-driven clinical decision support methods for early risk detection—such as machine learning and NLP approaches for identifying acute coronary syndrome from clinical records—helping clinicians prioritize patients and intervene earlier. At the same time, our healthcare research emphasizes practical, deployable tools that improve accessibility and patient-centered care. This includes predictive frameworks that help anticipate the scope needs of outpatient visits (e.g., otolaryngology clinics), as well as systematic, evidence-based work on diabetes management—ranging from optimization and decision-making approaches for medication prioritization to comprehensive reviews on machine learning for affordable and accessible insulin management. Finally, as extended reality becomes more common in training and clinical workflows, we also study how to select appropriate XR systems for healthcare using structured multi-criteria decision-making methods. Together, these efforts highlight how modern analytics can improve patient outcomes, strengthen operational resilience, and enable more efficient, equitable healthcare delivery.