Modeling and visualization of multimodal brain monitoring in acute brain injury
The intensive care medicine department (ICU) treat patients in life threating conditions, who usually require multi-organs support. ICU combines high clinical skills, a fully dedicated patient care system and sophisticated technology. Moreover, different physiological parameters are continuously monitored. Despite that, the ICU mortality in Europe remains around 20%. Acute brain injuries (ABI) gather most of neurological injuries that occur in ICU. ABI are not hereditary or congenital and include pathologies as: traumatic brain injury (TBI), strokes, brain tumors or other pathological status acquired after birth. The mechanisms that rule some ABI are not completely understood. Moreover, the initial moment in which very aggressive treatments should be executed is not well defined yet. In this complex landscape engineering techniques could help to better understand both the physiology and the best treatment approach for most of ABI injuries. My project focuses on the application of data mining, machine learning and signal processing techniques on ABI. The aim is to investigate the physiology and the hidden mechanisms behind these injuries in ICU. The project expects to extract and integrating knowledge from different sources, in detail, physiological signals, clinical measurements and categorical data are processed and analyzed. The aim is reducing data overload and making predictive modelling of ABI occurrences and outcomes. The final ambitious goal of the project is providing to ICU clinicians a better overview of ABI, providing new instruments to improve the current treatments.