Correlation between injury pattern and Finite Element analysis in biomechanical reconstructions of Traumatic Brain Injuries

J Biomech. 2015 May 1;48(7):1331-5. doi: 10.1016/j.jbiomech.2015.02.057. Epub 2015 Mar 12.

Abstract

At present, Finite Element (FE) analyses are often used as a tool to better understand the mechanisms of head injury. Previously, these models have been compared to cadaver experiments, with the next step under development being accident reconstructions. Thus far, the main focus has been on deriving an injury threshold and little effort has been put into correlating the documented injury location with the response displayed by the FE model. Therefore, the purpose of this study was to introduce a novel image correlation method that compares the response of the FE model with medical images. The injuries shown on the medical images were compared to the strain pattern in the FE model and evaluated by two indices; the Overlap Index (OI) and the Location Index (LI). As the name suggests, OI measures the area which indicates both injury in the medical images and high strain values in the FE images. LI evaluates the difference in center of mass in the medical and FE images. A perfect match would give an OI and LI equal to 1. This method was applied to three bicycle accident reconstructions. The reconstructions gave an average OI between 0.01 and 0.19 for the three cases and between 0.39 and 0.88 for LI. Performing injury reconstructions are a challenge as the information from the accidents often is uncertain. The suggested method evaluates the response in an objective way which can be used in future injury reconstruction studies.

Keywords: Accident reconstructions; Bicycle accidents; FEA; Head injuries; Image correlation.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Accidents*
  • Aged
  • Bicycling
  • Biomechanical Phenomena
  • Brain Injuries / physiopathology*
  • Computer Simulation
  • Craniocerebral Trauma
  • Female
  • Finite Element Analysis
  • Humans
  • Male
  • Middle Aged
  • Models, Anatomic
  • Models, Theoretical
  • Probability