The IBM Data Science and AI Elite helps Geisinger break new ground in sepsis care

By | 3 minute read | May 3, 2019

Sepsis. It’s a life-threatening infection that kills nearly 270,000 patients in the United States. Every year. What’s shocking is that one sepsis patient dies every two minutes – and 80 percent of these deaths are preventable by early detection. Not only is sepsis deadly, it’s also the most expensive condition to treat in the U.S., costing the nation’s more than $24 billion in hospital expenses annually.

Ricardo Balduino, IBM Data Science and AI Elite team
Anna Hazard, IBM Data Science and AI Elite team

It’s true that anyone can get an infection—and any infection can lead to sepsis. However, understanding which sepsis patients are at the highest risk for death can greatly help clinicians prioritize care. But up until now, the process of diagnosing and accurately making that prediction has been difficult.

That’s why Geisinger Health System, based in Danville, Penn., aimed to use data science to identify patients more quickly and accurately to ensure proper treatment and help decrease mortality rates. To do that, the Geisinger team set out to build a predictive model for sepsis mortality based on data from an actual, critical hospital setting. But their group of three, comprised of scientists and clinicians, included no data scientists. So they turned to the IBM Data Science and AI Elite team.

The IBM Data Science and AI Elite assembled a six-person swat team to develop two projects: a model to predict sepsis mortality, and a tool to keep the team on top of the latest sepsis research.

Part 1: Building a model to predict sepsis mortality with Watson Studio

For the first part of project, the team used IBM Watson Studio open source tools to build a predictive model that would ingest clinical data from thousands of sepsis patients spanning a decade. Then they used all the data to build another model to predict patient mortality during the hospitalization period or during the 90 days following their hospital stay.

The predictive model helped researchers identify clinical biomarkers associated with the higher rates of mortality from sepsis by predicting death or survival of patients in the test data.

Avijit Chatterjee, IBM Data Science and AI Elite team
Brittany Bogle, IBM Data Science and AI Elite team

Armed with the new model, Geisinger will be able to develop more personalized clinical care plans for at-risk sepsis patients. Potentially Geisinger will be able to increase patient chances of recovery by paying attention to the key factors linked to sepsis deaths—including age, the number of hospital transfers and time spent on vasopressor medicines.

Part 2: Using Watson Explorer to make it easier for Geisinger to gather the latest sepsis research

Around two million new medical journal papers emerge every year. Geisinger was looking for way to help keep their researchers apprised of all recent and archived studies. For this part of the engagement, the IBM Data Science and AI Elite team used IBM Watson Explorer to create a searchable index of thousands of medical publications. Clinicians and researchers can now mine the journal archive to uncover the most relevant content within a few clicks.

For Dr. Hosam Farag, a post-doctorate fellow at Geisinger Medical Institute, machine learning was a tool the team had not utilized. But the step-by-step process of the Elite team helped Geisinger understand the critical value of data science.

Vinay Rao Dandin, IBM Data Science and AI Elite team

“It’s very important for me as a clinician and a research scientist to save patient lives using all the knowledge of the data and the clinical backgrounds,” he said. “Machine learning can close the care gaps and optimize the treatment. That make me passionate about how to save patient lives. “

“Predicting all-cause death in sepsis patients can guide health providers actively monitor and take preventive actions to improve patients’ survival,” said Richard Balduino, solution architect with the IBM Data Science and AI Elite team. “Many of the features that were identified as important in our model are known to be associated with sepsis patients’ death. This provides reassurance that our machine learning models can help identify well-known associations with sepsis death even among the noise of many unrelated variables.”

Want to learn how machine learning and AI are transforming all kinds of business? Follow IBM clients throughout their journey to AI in our new client gallery and learn who were among the first to confidently put AI to work in their industry.

Find out how the IBM Data Science and AI Elite team can help you build out your AI strategy.

Accelerate your journey to AI with a prescriptive approach. Visit to learn how to modernize, collect, organize, analyze and infuse all your data.