Brain injury in critical illness

Important decisions regarding patient treatment in the ICU or Emergency Room are often made using only physical indicators and patient history at time of admission. With a wide range of treatment options, this initial assessment plays an important role in determining course of action.

Access to previous medical records including vital trends prior to and post injury or medical incident are used by our lab to model and predict the incidence and severity of injury with the goal of incorporating these algorithms into real-time diagnosis for better accuracy and treatment.

Children who are admitted to the hospital with decreased level of consciousness after a traumatic brain injury (TBI) require urgent medical attention. Treatment will depend on the type and severity of injury. Unfortunately, history and physical findings are often unreliable in the first hours of hospitalization, which is the critical time period in which urgent management decisions must be made.

We have developed a promising tool for measuring detectable evidence of TBI on routine brain scans. The tool combines features invisible to the human eye but detectable by computer software with expert knowledge. We want to evaluate how well our tool can perform in a real health care setting. We believe it will greatly improve the efficacy and quality of care provided to children after TBI.

Lab members involved in the study:

• Anna Colucci
• Geraldine Goco
• Helena Frndova
• Matthew Van Huyse
• Kyla McDonald

This observational study looks to identify patterns in NIRS and non-invasive cerebral signals prior to cerebral hemorrhaging in 27 weeks preterm babies. After parental consent, infants undergo continuous cerebral monitoring using NIRS (cerebrals Sao2) and amplitude integrated EEG for first 72 hours after birth together with serial echocardiograms and head ultrasounds during this period. Understanding early predictive markers of cerebral hemorrhages will help identify infants at risk of developing hemorrhages and those who will benefit from preventative strategies. Preventing cerebral hemorrhages will ultimately improve long term neuro-developmental outcome for these infants.

Lab members involved in the study:

• Poorva Deshpande
• Julia Dirks

RiEEG’s goal is to develop a program that can monitor brain health all the time without invasive procedures at the bedside for critically ill children. Brain waves will be recorded from the child’s scalp and evaluated using computer science techniques, like signal processing and machine learning. The final product is ideally an easy-to-use brain monitoring device that can analyze the brain patterns of ill children to detect changes in function in real time. This will ultimately shorten the window of time between injury and treatment.

Lab members involved in the study:

• Rishi Lalgudi Ganesan