Computational Surgery: Helping Surgeons Avoid Mistakes with Better Robots

Abstract

Preventable medical errors are the third leading cause of death in the United States. Despite over a decade of clinician-led efforts in policy and education, this issue remains. In the meantime, hospitals have adopted surgical robots at a dramatic pace. This provides opportunities to augment the art of surgery with more rigorous, quantitative science. This gives rise to the field of computational surgery which promises to address long-standing challenges in healthcare like the prevalence of human error. This talk will focus on two research problems in this area. First, how do we quantify and improve the existing skills of a surgeon? This requires a method whose scores correlate with patient outcomes, that can scale to cope with 51 million annual surgeries in the United States, and that can generalize across the diversity surgical procedures or specialties. Second, how can we build new robotic tools that render surgical tasks fundamentally easier, perhaps making errors impossible in the first place? This will survey multiple topics such as policy-blended human-robot shared control to ensure safety in robotic tissue grasping; novel patient-specific catheter robots that safely remove plaque via inverse design of soft robots and a theranostic excimer laser; and robotic 3D bioprinting directly onto moving human anatomy to explore new reconstructive procedures.