Tiffany Kashima

My experience in healthcare and medicine began as a bedside nurse and educator in pediatric critical care, congenital heart surgery, and ECMO.  After observing the reductionist modeling and frameworks employed across the dimensions of care delivery and instinctually wanting something that better captured the nuance and context of actual practice, I began an academic journey to study data science applied mathematics, and complex adaptive systems (complexity science).

Using the theory and frameworks of complexity science, I am working on the ground level with clinicians, allied health staff, and researchers to bring real world feedback into the modeling and frameworks that better prepare our clinicians to make decisions based on the most critical points in the dynamics of acute medicine, from communication patterns to clinical heuristics.  The goal is a realistic one that comes from being the "ground" in acute care myself-- improve decision making in meaningful ways that align with patient outcomes without ignoring the messy reality in which we practice-- reimbursement models, resource limitations, administrative and business facing interests, patient perceptions of health and medicine, etc.  Complexity makes it possible to capture the essential algorithms or interactions for emergent states at different levels in a system.  While most interventions are traditionally seen percolating from the top-down, I propose that the most influential ones are developed and implemented at the actual clinician level, both in practice and in education.

I have additional special interest in machine learning models that create realistic sandbox environments that support decision making and high fidelity learning. I work with engineers to guide them in asking the right questions about data, and in reconciling predictive power to the medical ethics that require explainability.