This abstract has been withheld from publication.

Wildfire smoke is becoming an increasingly common and persistent threat to public health, often affecting populations far outside of the range of active fires. In clinical settings, patients rarely report smoke exposure itself, instead presenting with symptoms such as shortness of breath, chest discomfort, impaired concentration, or ongoing psychological stress. This project examines wildfire smoke as a multisystem exposure, with evidence demonstrating impacts across respiratory, cardiovascular, and neurologic domains. Fine particulate matter (PM2.5) is the primary driver of these effects, as it penetrates deep into the lungs and initiates both local and systemic inflammatory responses. Smoke from fires involving built environments may further increase toxicity through the presence of metals and reactive compounds. The burden of these health effects is unevenly distributed, as biologically vulnerable groups; including children, older adults, and those with preexisting cardiopulmonary conditions; face elevated risk, while social factors such as reduced access to healthcare constrain the ability to mitigate exposure. While respiratory outcomes remain the most well-established, emerging research links wildfire smoke to cardiovascular events as well as cognitive disruptions. Overall, this work supports a clinical approach that frames wildfire smoke as a recurring, complex exposure requiring alignment of patient care with broader public health strategies.

Background: Simulation is essential in healthcare education, but high-fidelity access is limited by cost, faculty, space, and scheduling. Traditional simulations also limit repetition, realism, and individualized feedback. AI now makes immersive scenarios faster and cheaper to produce. Purpose/Aims: To develop and pilot an AI-driven simulation platform that improves clinical judgment and decision-making through immersive learning outside traditional simulation labs. Methods: Simtura.ai is a web-based platform that creates first-person clinical scenarios aligned with scope-of-practice competencies. Learners make assessment and intervention choices, and the system adapts patient responses based on those decisions. Results: Early testing showed high engagement, realism, and increased confidence in prioritization and rapid assessment. Participants reported better ability to sequence primary assessments and anticipate deterioration. Implications: This scalable model supports workforce readiness, reinforces evidence-based practice, and complements traditional simulation.