Tools and Processes for Assessment of Blast Overpressure–Related Traumatic Brain Injury

Author(s):

Dale Shires, Joseph Zirilli, Michael An, Robert Olsen, Michael Amato, Jacob Adams, and Katherine Breczinski

Abstract:

The blast overpressure studies (BOS) effort is attempting to understand and mitigate the effects of blast overpressure events for the military, and in particular address issues related to traumatic brain injury. The goal of this particular project was to better understand data analysis required to improve operational medicine for the military in terms of BOS. Also of interest was how this data should be cataloged and made available to researchers pursuing this topic. Assessed data generally consists of concentrations of blood biomarkers present, Defense Automated Neurobehavioral Assessment metrics, and self-reported symptoms by the participants. Several Python tools were used to reproduce the analyses present in the literature, including Spearman’s rank correlation, analysis of variance, and interquartile range assessments. Additional methods of analysis common in machine learning were also explored to see if useful correlations could be found. This work shows the advantages of having a structured approach for data storage and access, as well as highlighting the ability for common open source tools such as Python and various statistical and machine learning packages to produce meaningful results for analysis.