Hi Louise, tell us a bit about yourself. What sparked your interest in smart health?

I was trained as a biomedical engineer and am a mathematician and programmer at heart. During residency, I pursued a Masters degree in Epidemiology and Biostatistics, which enabled me to integrate my interests in data modeling and software engineering to create clinical solutions in real-time.

Where do you see AI making the greatest impact in anaesthesiology?

Remote monitoring and personalized predictive modeling are the future of anesthesiology. There’s a great wealth of routinely collected data in anesthesiology that’s not often available in other clinical specialties. These data could be harnessed to improve patient outcomes by forecasting critical events and providing personalized risk stratification to inform patient-centered decision-making. Automated closed-loop systems is another area for development.

How have you managed perceptions around trust of artificial intelligence – both within the hospital and with patients?

In working closely with patients, clinicians, and administrators, I realized that making AI explainable is the way to go in healthcare. It is always easier to trust in something that one could grasp. Acting as the intersection between IT, data scientists, clinicians, clinical leadership, and patients to ensure close communication of the science, needs, and concerns, has been a very rewarding aspect of my work.

Collaboration between clinicians, scientists and engineers is vital to the success of any health/data science project. Do you have any tips on how to bring cross-disciplinary and cross-functional teams together?

Yes! This is where a DIY approach works well for clinician data scientists. Communicate well, make all team members feel valued by giving recognition to their individual strengths, exercising patience in bridging mutual knowledge gaps, and promoting success by being the “bridge” that integrates each member of the interdisciplinary team.

In what ways do you see AI shaping the future of precision medicine?

I see AI as a means of integrating a myriad of routinely collected data sources to provide personalized digital signatures, which can in turn be used to provide disease surveillance, risk stratification, and treatment in a way that is seamless.

We believe in changing healthcare one connection at a time. If you are interested in the opinions in this piece, in connecting with the author, or the opportunity to submit an article, let us know. We love to help bring people together! [email protected] 

Dr. Sun is Professor and Chief of Cardiothoracic Anesthesiology at Stanford. Her areas of clinical focus are hemodynamic monitoring and heart failure. Her methodologic areas of focus are the conduct of population-based cohort studies, predictive analytics, sex and gender epidemiology, patient engagement, data warehousing, and applications development. Her patient-centered research program leverages big data and digital technology to bridge key gaps in the delivery of care and outcomes for patients with heart failure and those undergoing cardiovascular interventions, through personalized risk stratification and characterizing long-term, patient-defined outcomes. She specializes in rapidly developing and deploying data-driven solutions to enhance clinical operations and patient care, and collaborates with policy makers to evaluate models of cardiac healthcare delivery. Dr. Sun sits on a number of editorial boards and scientific review committees internationally. She has authored over 100 peer-reviewed publications, many in leading journals including JAMA, JAMA Cardiology, JAMA Internal Medicine, Circulation, JACC, and Diabetes Care.