My role as a data scientist with the IBM Data Science Elite Team gives me great opportunities to work in the field with client organizations doing important work in healthcare and medical science. During Think 2019, February 12–15, I’ll join two physicians to discuss some of the work we’ve been able to do together.

First, on Wednesday, February 13th, at 8:30 AM, I’ll be presenting on Predicting Outcomes of Patients with Sepsis Using Machine Learning (Session ID: 2605A). My co-presenter will be Dr. Hosam Farag (MD, MSPH), a research scientist at Geisinger Diagnostic Medicine Institute with a passion for using data science to better diagnose patients suffering from sepsis, a disease caused when the body has an extreme response to infection.

This work is so important because sepsis has a high mortality rate when diagnosed late, but is highly treatable if diagnosed in time. Through our partnership with Geisinger Health System, we started with historical clinical data about sepsis patients, then applied machine learning techniques in IBM Watson Studio to create a model for predicting which patients are more susceptible to fatal outcomes from any cause, either in the hospital or after discharge.

Second, on Thursday, February 14th, at 9:30 AM, I’ll join Dr. Farag and Dr. Shravan Kethireddy (MD) for Finding Insights in Scientific Publications Using Machine Learning and Natural Language Processing (Session ID: 2525A). Dr. Kethireddy, formerly Associate Chief Data Officer at Geisinger Medical Center, is a board-certified physician who is now the System Medical Director of Critical Care Medicine at the Northeast Georgia Health System.

This session will explain how IBM is helping health professionals and researchers to navigate and quickly find new insights from the many scientific publications available today. To address this problem, we used natural language processing and machine learning techniques and tools such as IBM Watson Explorer and IBM Watson Studio to ingest, parse, and browse a large number of medical publications, and made use of powerful queries and visualizations to gain insights and find publications of interest faster.

Both sessions will be held in San Francisco Ballroom 214 on Level 2 of the Moscone Center. If you’d like to hear more about how data science is having a real impact on healthcare, I hope to see you there.