During my pulmonary fellowship, I encountered a thoracic surgeon who boldly declared, "Whenever I see a pre-op chest x-ray with coronary calcifications, I worry about complications. So, I do extra cardiac work-ups on those patients." Inwardly, I rolled my eyes and labeled his anecdotal correlations as "totally unscientific." Two decades later, I found myself smiling while reading the 2010 American Heart Association / American College of Cardiology recommendation on CT coronary artery calcification scoring for asymptomatic patients at intermediate CVD risk. That surgeon had made a connection where I saw none.
What if a brain bigger than yours or mine could discover meaningful correlations that could improve care? What if uncertainty in decision-making could be addressed by innovations in technology? IBM supercomputer Watson might be up for the challenge. After besting human champions on Jeopardy! in 2011, Watson went to medical school, using natural language processing to consume vast volumes of medical literature and clinical records, with the ability to synthesize the career experience of thousands of clinicians across various specialties. Submit clinical case details to the pizza-box-sized computer and it will return statistical probabilities associated with various courses of action, along with supporting evidence.
Guided by medical experts, Watson has lately been specializing in cancer care to support complex clinical decision-making, and can generate a synthesis of big data unaffected by personal bias. Through rounds of feedback from clinicians, the supercomputer refines its analysis of predicted clinical outcomes.
"One unique aspect of the MD Anderson Oncology Expert Advisor, powered by Watson, is that it will not solely rely on established cancer care pathways to recommend appropriate treatment options," explains Lynda Chin, M.D., professor and chair of Genomic Medicine and scientific director at the Institute for Applied Cancer Science at the University of Texas’ MD Anderson Center. "The system was built with the understanding that what we know today will not be enough for many patients."
Like Watson, at Epocrates, we synthesize relevant data, big and small, to support the decisions clinicians make in the moments of care. In our nascent years, Epocrates began by distilling vast volumes of drug knowledge into succinct, actionable guidance that informs clinical decisions. Last year, we released our antibiotic resistance app, Epocrates Bugs + Drugs, a mobile decision support tool with bacterial susceptibility data localized down to the ZIP code. Its innovative algorithm filters thousands of microbiology lab data points taken from records on the cloud-based athenahealth network each day, and serves up resistance data relevant to patients living in any U.S. community.
So, what’s next? Clinicians tell us that clinical practice guidelines from major specialty societies are highly valuable assets to support decision-making. But there’s a hurdle: Many of the specialty guidelines exceed 80 pages, and are not optimized for rapid assimilation. This year, we plan to filter guidelines for clinicians, extracting the actionable content to bring more practice-changing decision support to moments of care. We are inspired by our clinician-advisors and technology like Watson that is redefining patient care. Both remind us that we’re on the right track, helping highlight the relevant, one data point at a time.
Obsessed with British television, I assumed that Watson was named after Sherlock's trusted physician-companion. It's not - nor is it named after the famous DNA pioneer James Watson. The name is a reference to Thomas Watson, IBM's founder.