Resilience in CCI

Chronic critical illness (CCI) refers to a state where ICU patients, after surviving a severe initial event, remain dependent on prolonged intensive care. Excluding severely injured patients, it is challenging to anticipate patient recovery in advance. This is mainly because the signals leading up to patient deterioration are quite weak. In this project we seek to overcome this limitation by combining machine learning and complex systems modeling. Using data from a Trauma ICU from an academic hospital, we model the patient as a complex adaptive system. By integrating machine learning models with complex systems perspective, patient resilience is measured via dynamic indicators of resilience, signals for critical slowing down are extracted to measure precursors of patient deterioration. We explore how taking a complex systems perspective has implications for our understanding of patient condition, possibilities of assessment and patient recovery.

Talks

• Muhammad Aurangzeb Ahmad Integrating Machine Learning and Complex Systems perspectives on Chronic critical illness at the Air Force Institute of Technology Dayton, OH August 20, 2024

Team Members

• Grant O’Keefe M.D.
• Norma Elzinga