We propose a methodology to develop a “smart” sepsis prognosis prediction algorithm that is linked to best clinical practice interventions by combining knowledge of sepsis prognosis learned from electronic health record data and clinician’s perspective.
Based on the developed algorithm, we can define CDS specification that can be implemented in any EHR system. This will help clinicians to facilitate early detection of sepsis recurrence and improve sepsis management for hospitalized patients.

Describe the new knowledge and additional skills the participant will gain after attending your presentation.: 1. Learning new clinical knowledge from electronic health record data set
2. Understading how to build a precise prediction model by combining structured data (e.g., lab, medication, and vital sign data) and unstructured clinical documentation


Min Jeoung Kang (Presenter)
Brigham and Women's Hospital

Li Zhou, Brigham and Women's Hospital
Frank Chang, Brigham and Women's Hospital
Christopher Knaplund, Columbia University, School of Nursing
Jose Garcia, Brigham and Women's Hospital
Kenrick Cato, Columbia University, School of Nursing
Sarah Collins, Columbia University, School of Nursing
Patricia Dykes, Brigham and Women's Hospital

Presentation Materials: